# 302.AI API Document ## Docs - 根目录 > Large Language Model [API Migration Guide](https://doc-en.302.ai/5030894m0.md): - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox [Introduction to Claude Code Sandbox](https://doc-en.302.ai/7892894m0.md): - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox [Third-Party Client Access Tutorial](https://doc-en.302.ai/7892896m0.md): - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox [Update Log](https://doc-en.302.ai/7892899m0.md): - 根目录 > Image Generation > Unified interface [Explanation ](https://doc-en.302.ai/7062445m0.md): - 根目录 > Video Generation > Unified Interface [Explanation ](https://doc-en.302.ai/7059949m0.md): - 根目录 > Video Generation > PixVerse [Pixverse Special Effect ID](https://doc-en.302.ai/6668087m0.md): - 根目录 > Video Generation > PixVerse [Pixverse Sound Effect ID](https://doc-en.302.ai/6668103m0.md): - 根目录 > Video Generation > Jimeng [seedance Video Generation Price Reference](https://doc-en.302.ai/7948566m0.md): - 根目录 > Video Generation > Higgsfield [Image-to-Video Template](https://doc-en.302.ai/7350262m0.md): - 根目录 > Information Processing > 302.AI > Code execution > E2B SDK Invocation [E2B SDK](https://doc-en.302.ai/7437521m0.md): - 根目录 > Tools API > Pay with 302 [Obtain the signature algorithm](https://doc-en.302.ai/7669570m0.md): - 根目录 > Tools API > AI PPT Generator [Instructions for Using AI PPT Interface](https://doc-en.302.ai/8021425m0.md): - 根目录 > Tools API > One-Click Website Deployment [Upadte ](https://doc-en.302.ai/7524491m0.md): - 根目录 > Tools API > AI Image Creative Station API [nable asynchronous request example (It is recommended to use asynchronous requests to generate images. Image generation models take a long time, and using asynchronous mode can help reduce request timeout issues.)](https://doc-en.302.ai/6722676m0.md): - 根目录 > Tools API > AI Image Creative Station API [List of models that support image generation](https://doc-en.302.ai/7783560m0.md): - 根目录 > Help Center [HTTP Status Codes](https://doc-en.302.ai/5030895m0.md): - 根目录 > Help Center [List of supported languages for image translation](https://doc-en.302.ai/5213777m0.md): ## API Docs - 根目录 > Large Language Model > Exclusive Feature > Search Online [Chat（Search online）](https://doc-en.302.ai/273308610e0.md): Add internet search capability to all models. There are two ways to enable this feature, and you can choose either one: - 根目录 > Large Language Model > Exclusive Feature > Depth-First Search [Chat（Depth-First Search）](https://doc-en.302.ai/270155422e0.md): Enhance deep search capabilities for all models: - 根目录 > Large Language Model > Exclusive Feature > Image Analysis [Chat（Image analysis）](https://doc-en.302.ai/260156326e0.md): adds image recognition capabilities to all models, and there are two ways to enable it, you can choose either one: - 根目录 > Large Language Model > Exclusive Feature > Reasoning mode [Chat（Reasoning mode）](https://doc-en.302.ai/266354843e0.md): Add inference ability to all models, just add the suffix -r1-fusion to the model name. - 根目录 > Large Language Model > Exclusive Feature > Link Parsing [Chat（Link Parsing）](https://doc-en.302.ai/272503066e0.md): Add webpage/file parsing capabilities to all models, with two available methods—choose either one: - 根目录 > Large Language Model > Exclusive Feature > Tool Invocation [Chat（tool invocation）](https://doc-en.302.ai/278668549e0.md): The 302 platform has added tool invocation capabilities (commonly known as Function Call) to all models. - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > User Management [Create User](https://doc-en.302.ai/255988308e0.md): In the Memobase system, create a new user - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > User Management [Get User](https://doc-en.302.ai/255988309e0.md): In the Memobase system, obtain a user's information - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > User Management [Update User](https://doc-en.302.ai/255988310e0.md): In the Memobase system, update a user's information - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > User Management [Delete User](https://doc-en.302.ai/255988311e0.md): In the Memobase system, delete existing users - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Data Management [Insert Data](https://doc-en.302.ai/255988312e0.md): In the Memobase system, insert short-term memory data for a user - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Data Management [Get Datas](https://doc-en.302.ai/255988313e0.md): In the Memobase system, obtain a user's short-term memory data list - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Data Management [Get Data](https://doc-en.302.ai/255988314e0.md): In the Memobase system, obtain a user's short-term memory data - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Data Management [Delete Data](https://doc-en.302.ai/255988315e0.md): In the Memobase system, delete a user's short-term memory data - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Memory Management [Flush Buffer (Generate Memory)](https://doc-en.302.ai/255988316e0.md): In the Memobase system, extract user short-term memory cache into long-term memory - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Memory Management [Get User Profile (Get Memory)](https://doc-en.302.ai/255988317e0.md): In the Memobase system, extract the corresponding user memory - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) > Memobase > Memory Management [Delete User Profile (Delete Memory)](https://doc-en.302.ai/255988318e0.md): In the Memobase system, delete the corresponding user memory - 根目录 > Large Language Model > Exclusive Feature > Long-term memory (Beta) [Chat (Long-term Memory)](https://doc-en.302.ai/255482146e0.md): To add long-term memory functionality to any large model, simply include a `userid` parameter. - 根目录 > Large Language Model > Exclusive Feature > Simplified API [Chat (Simplified API)](https://doc-en.302.ai/207705101e0.md): We provide a simplified API version, where you only need to pass model and message to get the output. Suitable for scenarios where streaming output is not required and only the result is needed. - 根目录 > Large Language Model > Exclusive Feature > Asynchronous call [Asynchronous request to chat](https://doc-en.302.ai/323087731e0.md): The existing /chat/completions endpoints all support asynchronous responses—just add the parameter async=true to your query. This feature is provided at no extra cost. We have also added a callback parameter to support webhooks; note that callbacks require async=true. - 根目录 > Large Language Model > Exclusive Feature > Asynchronous call [Asynchronously retrieve/get results](https://doc-en.302.ai/323084919e0.md): **Price: 0 PTC/time** - 根目录 > Large Language Model > Exclusive Feature > Claude Format [Messages（Claude Format）](https://doc-en.302.ai/326799367e0.md): For better compatibility with Claude Code, we now support invoking all models using the Claude format. - 根目录 > Large Language Model > Claude Code > Claude Model 30% Discount Area [Claude Code Dedicated URL Path](https://doc-en.302.ai/394284007e0.md): Only in Claude Code, change the API Base to: - 根目录 > Large Language Model > Claude Code > Claude Model 30% Discount Area [Claude Code Dedicated Model Name](https://doc-en.302.ai/394284008e0.md): If you don't want to use the Claude Code dedicated URL Path, you can also change the model name. Our currently supported 30% off models are: - 根目录 > Large Language Model > Claude Code > 30% Discount Area for Chinese-made Models [glm-for-coding (Zhipu)](https://doc-en.302.ai/394326369e0.md): On 302.AI, you can use **glm-for-coding** to enjoy a 30% discount on Zhipu models. - 根目录 > Large Language Model > Claude Code > 30% Discount Area for Chinese-made Models [kimi-for-coding（Moonlight）](https://doc-en.302.ai/394326370e0.md): On 302.AI, you can use **kimi-for-coding** to enjoy a 30% discount on Moonlight models. - 根目录 > Large Language Model > Claude Code > Any model is compatible with Claude format [Messages (Claude Format)](https://doc-en.302.ai/394335181e0.md): For better adaptation to Claude Code, we support all models to be called using the Claude format. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Create Claude-Code Sandbox](https://doc-en.302.ai/394338638e0.md): ## Preinstalled Environment - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Query the sandbox list under the ApiKey](https://doc-en.302.ai/394340724e0.md): **Price: Free** - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Delete Sandbox](https://doc-en.302.ai/394340725e0.md): All data within the sandbox will be unrecoverable after deletion. Please download and save the files yourself before deletion. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Update sandbox configuration](https://doc-en.302.ai/394341730e0.md): **Price: Free** - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Add MCP to the sandbox](https://doc-en.302.ai/394341733e0.md): **Price**: 0.0005 PTC * seconds of sandbox usage - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Query all sessions of the sandbox](https://doc-en.302.ai/394341734e0.md): **Price**: Free - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Add/modify conversation notes](https://doc-en.302.ai/394341735e0.md): **Price**: Free - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Management [Delete the specified session and all files in the corresponding working directory](https://doc-en.302.ai/394341736e0.md): **Price**: 0.0005 PTC * seconds of sandbox usage - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Interaction [Call Claude-Code (OpenAI format)](https://doc-en.302.ai/394338639e0.md): Using Claude Code through the OpenAI language model interface format **only supports stream responses**. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Interaction [Execute instructions using prompt (OpenAI format)](https://doc-en.302.ai/394338640e0.md): Supports inputting instructions in user prompts to execute specific functions, including operations such as executing commands, deploying applications, and changing the LLM model of the sandbox. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Interaction [Call Claude-Code (Claude format)](https://doc-en.302.ai/394338641e0.md): Using Claude Code through the Anthropic language model interface format **only supports stream responses**. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Interaction [Execute instructions using prompt words (Claude format)](https://doc-en.302.ai/394338642e0.md): Supports inputting instructions in user prompts to execute specific functions, including operations such as executing commands, deploying applications, and changing the LLM model of the sandbox. - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > Sandbox Interaction [Execute the sandbox command instruction](https://doc-en.302.ai/394341731e0.md): - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [View the file list under the sandbox](https://doc-en.302.ai/394340726e0.md): View file information inside the sandbox - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [Download file (supports batch download)](https://doc-en.302.ai/394341726e0.md): ## Function Description - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [Upload files to the sandbox (JSON)](https://doc-en.302.ai/394341727e0.md): Supports direct file upload via form-data && supports file upload in JSON format via url/base64 with header information (data: xxxx) - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [Upload files to the inside of the sandbox (from-data)](https://doc-en.302.ai/394341728e0.md): Supports direct file upload via form-data && supports file upload in json format via url/base64 with header information (data: xxxx) - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [File operation interface](https://doc-en.302.ai/394341732e0.md): **Price**: 0.0005 PTC * seconds of sandbox usage - 根目录 > Large Language Model > Claude Code > Claude Code Sandbox > File-related [Deploy the project](https://doc-en.302.ai/394341729e0.md): Deploy the generated project - 根目录 > Large Language Model > Model Support [Models (List models)](https://doc-en.302.ai/261684084e0.md): List the currently available models and provide prices for each model. - 根目录 > Large Language Model > Model Support [Status（Model Status）](https://doc-en.302.ai/284770652e0.md): We monitor the first-token response time for some models. You can check the model service availability through this interface. - 根目录 > Large Language Model > OpenAI [Chat（Talk）](https://doc-en.302.ai/207705102e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Responses（Talk）](https://doc-en.302.ai/308965988e0.md): [Official Guide](https://platform.openai.com/docs/guides/responses-vs-chat-completions) - 根目录 > Large Language Model > OpenAI [Chat（Streamed return.）](https://doc-en.302.ai/239842865e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (Image Analysis)](https://doc-en.302.ai/207705107e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (Structured Output)](https://doc-en.302.ai/207705108e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (function call)](https://doc-en.302.ai/216495993e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (gpt-4-plus image analysis)](https://doc-en.302.ai/207705105e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (gpt-4-plus image generation)](https://doc-en.302.ai/207705106e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat（gpt-4o-image-generation modify image）](https://doc-en.302.ai/282181336e0.md): Model Name: gpt-4o-image-generation - 根目录 > Large Language Model > OpenAI [Chat (gpts model)](https://doc-en.302.ai/207705103e0.md): GPTs are customizable gpt-4 models launched by OpenAI, allowing users to design their own AI assistants for various scenarios, such as paper search, translation, code completion, image generation, etc. - 根目录 > Large Language Model > OpenAI [Chat (chatgpt-4o-latest)](https://doc-en.302.ai/207705104e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Large Language Model > OpenAI [Chat (o1 Series Model)](https://doc-en.302.ai/216495992e0.md): Support the model below： - 根目录 > Large Language Model > OpenAI [Chat (o3 Series Model)](https://doc-en.302.ai/258254135e0.md): Support o3-mini, o3-mini-2025-01-31 - 根目录 > Large Language Model > OpenAI [Chat（o4 Series）](https://doc-en.302.ai/296806205e0.md): Supports o4-mini, o4-mini-2025-04-16 - 根目录 > Large Language Model > OpenAI [Chat（gpt-4o audio model）](https://doc-en.302.ai/225241341e0.md): This example demonstrates how to use the gpt-4o-audio-preview model. - 根目录 > Large Language Model > OpenAI [Responses（Deep-Research）](https://doc-en.302.ai/315837721e0.md): [Official Guide](https://platform.openai.com/docs/guides/responses-vs-chat-completions) - 根目录 > Large Language Model > OpenAI [Responses（Codex Cli专用）](https://doc-en.302.ai/359995819e0.md): You only need to change the API Base in Codex Cli to: - 根目录 > Large Language Model > Anthropic [Chat（Talk）](https://doc-en.302.ai/207705109e0.md): 302.AI currently support: - 根目录 > Large Language Model > Anthropic [Chat（Analyze image）](https://doc-en.302.ai/207705110e0.md): 302.AI currently support: - 根目录 > Large Language Model > Anthropic [Chat（Function Call）](https://doc-en.302.ai/219124319e0.md): 302.AI currently support: - 根目录 > Large Language Model > Anthropic [Messages（Original format）](https://doc-en.302.ai/222603718e0.md): 302.AI currently support: - 根目录 > Large Language Model > Anthropic [Messages（Function Call）](https://doc-en.302.ai/226294287e0.md): 302.AI currently support: - 根目录 > Large Language Model > Anthropic [Messages（Thinking mode）](https://doc-en.302.ai/264179581e0.md): Now claude-3-7-sonnet-20250219 supports thinking mode, which can be opened through parameters - 根目录 > Large Language Model > Anthropic [Messages（128k output）](https://doc-en.302.ai/264179582e0.md): Now claude-3-7-sonnet-20250219 supports outputs up to 128k, which is more than 15 times longer than other Claude models. This ability to extend is especially effective for extended thinking use cases involving complex inference, rich code generation, and comprehensive content creation. - 根目录 > Large Language Model > Anthropic [Messages (for Claude Code)](https://doc-en.302.ai/350161038e0.md): You only need to change the API Base in Claude Code to: - 根目录 > Large Language Model > Gemini > Official Format [v1beta (Official Format - Chat)](https://doc-en.302.ai/344797828e0.md): Original format from Google Gemini - 根目录 > Large Language Model > Gemini > Official Format [v1beta (Official Format - Text-to-Image)](https://doc-en.302.ai/344797829e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Large Language Model > Gemini > Official Format [v1beta (Official Format - Image Editing)](https://doc-en.302.ai/344797830e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Large Language Model > Gemini > Official Format [v1beta (Official Format - Streaming)](https://doc-en.302.ai/379755078e0.md): Original format from Google Gemini - 根目录 > Large Language Model > Gemini [Chat（Talk）](https://doc-en.302.ai/207705111e0.md): supports the following models: - 根目录 > Large Language Model > Gemini [Chat（Analyze image）](https://doc-en.302.ai/224540599e0.md): supports the following models: - 根目录 > Large Language Model > Gemini [Chat（Image Generation）](https://doc-en.302.ai/274416667e0.md): The newest Gemini models support both text-to-image and image-to-image generation. Simply pass an image URL using the `image_url` parameter. - 根目录 > Large Language Model > China AI Model [Chat (Baidu ERNIE)](https://doc-en.302.ai/207705112e0.md): The latest AI model from Baidu - 根目录 > Large Language Model > China AI Model [Chat (Tongyi Qianwen)](https://doc-en.302.ai/207705113e0.md): Alibaba's latest AI model - 根目录 > Large Language Model > China AI Model [Chat (Tongyi Qianwen-VL)](https://doc-en.302.ai/207705114e0.md): Alibaba's latest AI model, multimodal - 根目录 > Large Language Model > China AI Model [Chat（Tongyi Qianwen-OCR）](https://doc-en.302.ai/242312995e0.md): The latest AI model from Alibaba, which is a large OCR recognition model trained based on Qwen-VL, aggregates various text-image recognition, analysis, and processing tasks into a unified model, providing powerful text-image recognition capabilities. - 根目录 > Large Language Model > China AI Model [Chat (Zhipu GLM)](https://doc-en.302.ai/207705115e0.md): Zhipu AI's latest AI model, from Tsinghua University - 根目录 > Large Language Model > China AI Model [Chat (Zhipu GLM Multimodal)](https://doc-en.302.ai/207705116e0.md): Zhipu AI's latest image recognition AI model, from Tsinghua University - 根目录 > Large Language Model > China AI Model [Chat (Baichuan AI)](https://doc-en.302.ai/207705117e0.md): Baichuan AI's model, from Sogou founder Wang Xiaochuan - 根目录 > Large Language Model > China AI Model [Chat (Moonshot AI)](https://doc-en.302.ai/207705118e0.md): Moonshot's latest AI model, also used by the Kimi application - 根目录 > Large Language Model > China AI Model [Chat (Moonshot AI-Vision)](https://doc-en.302.ai/255026507e0.md): Moonshot's latest AI model, also used by the Kimi application - 根目录 > Large Language Model > China AI Model [Chat (01.AI)](https://doc-en.302.ai/207705119e0.md): Latest AI model from 01.AI, created by former Google vice president Kai-Fu Lee. - 根目录 > Large Language Model > China AI Model [Chat (01.AI-VL)](https://doc-en.302.ai/207705120e0.md): Latest AI model from 01.AI, created by former Google vice president Kai-Fu Lee. - 根目录 > Large Language Model > China AI Model [Chat (DeepSeek)](https://doc-en.302.ai/207705121e0.md): From the well-known private equity giant Fantasy. - 根目录 > Large Language Model > China AI Model [Chat (DeepSeek-VL2)](https://doc-en.302.ai/246110346e0.md): DeepSeek's latest AI model is currently the most affordable domestically produced large language model, with prices as low as 1 RMB per 1M input tokens and 2 RMB per 1M output tokens. It is highly suitable for translation tasks and comes from the renowned private equity giant, Phantom Funds - 根目录 > Large Language Model > China AI Model [Chat (ByteDance Doubao)](https://doc-en.302.ai/207705122e0.md): ****ByteDance Doubao's latest AI model** - 根目录 > Large Language Model > China AI Model [Chat (ByteDance Doubao-Vision)](https://doc-en.302.ai/240583755e0.md): Byte Doubao's latest image recognition model - 根目录 > Large Language Model > China AI Model [Chat（ByteDance Doubao Image Generation）](https://doc-en.302.ai/275421475e0.md): We have modified Doubao's general_v2.1_L and seededit APIs to adapt to OpenAI's format, enabling image generation and image editing. - 根目录 > Large Language Model > China AI Model [Chat (Stepfun)](https://doc-en.302.ai/246448203e0.md): Stepfun's latest AI model - 根目录 > Large Language Model > China AI Model [Chat (Stepfun Multimodal)](https://doc-en.302.ai/207705123e0.md): **Stepfun's latest AI model** - 根目录 > Large Language Model > China AI Model [Chat (iFLYTEK Spark)](https://doc-en.302.ai/207705124e0.md): **Xunfei Spark's latest AI model** - 根目录 > Large Language Model > China AI Model [Chat (SenseTime)](https://doc-en.302.ai/207705125e0.md): **Latest AI model from SenseTime** - 根目录 > Large Language Model > China AI Model [Chat（Minimax）](https://doc-en.302.ai/240583947e0.md): Minimax's latest AI model - 根目录 > Large Language Model > China AI Model [Chat (Tencent Hunyuan)](https://doc-en.302.ai/207705127e0.md): **Tencent Hunyuan large-scale model** - 根目录 > Large Language Model > SiliconFlow [Chat（SiliconFlow）](https://doc-en.302.ai/252564719e0.md): Silicon Stream I has officially partnered with 302.AI to provide open-source model capabilities to all 302 users. - 根目录 > Large Language Model > PPIO [Chat（PPIO）](https://doc-en.302.ai/308032503e0.md): PPIO Paiou Cloud has officially partnered with 302.AI to provide open-source model access to all 302 users. - 根目录 > Large Language Model > SophNet [Chat（SophNet）](https://doc-en.302.ai/319727732e0.md): SophNet has officially partnered with 302.AI to provide open-source model integration for all 302 users. - 根目录 > Large Language Model > Open Source Model [Chat（LLaMA4）](https://doc-en.302.ai/293003230e0.md): Meta's latest open-source model, reportedly surpassing the previous generation 405B - 根目录 > Large Language Model > Open Source Model [Chat（LLaMA3.3）](https://doc-en.302.ai/207705128e0.md): Meta's latest open-source model, reportedly surpassing the previous generation 405B - 根目录 > Large Language Model > Open Source Model [Chat（LLaMA3.2 multimodal）](https://doc-en.302.ai/219126691e0.md): **Meta's latest open source model** - 根目录 > Large Language Model > Open Source Model [Chat（LLaMA3.1）](https://doc-en.302.ai/207705129e0.md): **Meta's latest open source model** - 根目录 > Large Language Model > Open Source Model [Chat（Mistral）](https://doc-en.302.ai/235469484e0.md): A chat model takes a series of messages as input and then returns model-generated messages as output. Although the chat format is designed to facilitate multi-turn conversations, it is equally useful for single-turn tasks without any dialogue. - 根目录 > Large Language Model > Open Source Model [Chat（Pixtral-Large-2411multimodal）](https://doc-en.302.ai/235464754e0.md): This example is to demonstrate how to use the pixtral-large-2411 model to analyze images. - 根目录 > Large Language Model > Open Source Model [Chat（Gemma-7B、Gemma-3-27b-it）](https://doc-en.302.ai/207705132e0.md): Chat models take a series of messages as input and return a model-generated message as output. While the chat format is designed to make multi-turn conversations easy, it's just as useful for single-turn tasks without any conversation. - 根目录 > Large Language Model > Open Source Model [Chat（Gemma2-9B）](https://doc-en.302.ai/207705135e0.md): Chat models take a series of messages as input and return a model-generated message as output. While the chat format is designed to make multi-turn conversations easy, it's just as useful for single-turn tasks without any conversation. - 根目录 > Large Language Model > Open Source Model [Chat（Command R+） ](https://doc-en.302.ai/207705133e0.md): Chat models take a series of messages as input and return a model-generated message as output. While the chat format is designed to make multi-turn conversations easy, it's just as useful for single-turn tasks without any conversation. - 根目录 > Large Language Model > Open Source Model [Chat（Qwen2）](https://doc-en.302.ai/207705134e0.md): Chat models take a series of messages as input and return a model-generated message as output. While the chat format is designed to make multi-turn conversations easy, it's just as useful for single-turn tasks without any conversation. - 根目录 > Large Language Model > Open Source Model [Chat（Qwen2.5）](https://doc-en.302.ai/217022578e0.md): Alibaba's latest open source model** - 根目录 > Large Language Model > Open Source Model [Chat（Qwen2.5-VL）](https://doc-en.302.ai/263728248e0.md): **Alibaba's latest open source model** - 根目录 > Large Language Model > Open Source Model [Chat（Qwen3）](https://doc-en.302.ai/291197842e0.md): Qwen3 does not currently support non-streaming output. - 根目录 > Large Language Model > Open Source Model [Chat（Llama-3.1-nemotron）](https://doc-en.302.ai/224540845e0.md): Nvidia's fine-tuned model, built on Llama-3.1, ranks just behind o1 in performance scores. - 根目录 > Large Language Model > Open Source Model [Chat（QwQ-32B、QwQ-Plus、QwQ-32B-Preview）](https://doc-en.302.ai/239086885e0.md): Alibaba's latest open-source model - 根目录 > Large Language Model > Open Source Model [Chat（LongCat-Flash-Chat）](https://doc-en.302.ai/346058139e0.md): ![](https://raw.githubusercontent.com/meituan-longcat/LongCat-Flash-Chat/main/figures/longcat_logo.svg) - 根目录 > Large Language Model > Open Source Model [Chat（KAT-Coder）](https://doc-en.302.ai/366936071e0.md): KAT-Coder is a high-performance AI programming large model launched by StreamLake, the intelligent computing brand under Kuaishou. It is specifically designed for autonomous coding tasks, aiming to improve development efficiency and empower developers. With powerful Agentic Coding capabilities, it can autonomously reason, plan, and execute tasks through large-scale agent reinforcement learning, efficiently completing function development, defect analysis, unit test generation, and more. KAT-Coder supports over 20 mainstream programming languages, such as C, C++, Java, Python, Go, JavaScript, TypeScript, Rust, and is widely applicable to various development scenarios. - 根目录 > Large Language Model > Expert Model [Chat（WiseDiag Medical Model）](https://doc-en.302.ai/261594739e0.md): Model from Zhidia Technology: https://wisediag.com/ - 根目录 > Large Language Model > Expert Model [Chat (Xuanyuan Financial Model)](https://doc-en.302.ai/207705137e0.md): Xuanyuan financial model - 根目录 > Large Language Model > Expert Model [Chat (Farui Legal Model)](https://doc-en.302.ai/207705138e0.md): Farui legal model - 根目录 > Large Language Model > Expert Model [Chat (Alibaba Math Model)](https://doc-en.302.ai/207705139e0.md): **Alibaba math model** - 根目录 > Large Language Model > Expert Model [Chat（Perplexity search）](https://doc-en.302.ai/214715975e0.md): Supported models: - 根目录 > Large Language Model > Expert Model [Chat（Alibaba Tongyi Translation Model）](https://doc-en.302.ai/327572454e0.md): ### Flagship Translation Model Based on Qwen3 - 根目录 > Large Language Model > Other Models [Chat（grok-3）](https://doc-en.302.ai/263685475e0.md): The latest model trained by Elon Musk's xAI - 根目录 > Large Language Model > Other Models [Chat（grok-2）](https://doc-en.302.ai/224540600e0.md): The newest model developed by Elon Musk's xAI. - 根目录 > Large Language Model > Other Models [Chat（grok-2-vision）](https://doc-en.302.ai/246122225e0.md): The newest model developed by Elon Musk's xAI. - 根目录 > Large Language Model > Other Models [Chat（Nova）](https://doc-en.302.ai/242303358e0.md): The newest model developed by Amazon - 根目录 > Large Language Model > Other Models [Chat（v0）](https://doc-en.302.ai/312727858e0.md): Latest large models from v0: - 根目录 > Large Language Model > Other Models [Chat (UniFuncs Deep Research)](https://doc-en.302.ai/329405860e0.md): From UniFuncs' latest Deep Research model - 根目录 > Large Language Model > Other Models [Async Get Result](https://doc-en.302.ai/329405861e0.md): **Price: 0 PTC/time** - 根目录 > Image Generation > Unified interface > 302 Format V1 [Image Generation (302 Format)](https://doc-en.302.ai/286781078e0.md): Supports parameter passing in both application/json and multipart/form-data formats - 根目录 > Image Generation > Unified interface > 302 Format V2 [Synchronous Image Generation (302 Format)](https://doc-en.302.ai/340677693e0.md): Supports parameter passing in both application/json and multipart/form-data formats - 根目录 > Image Generation > Unified interface > 302 Format V2 [Asynchronous Image Generation (302 Format)](https://doc-en.302.ai/340677694e0.md): Supports parameter passing in both application/json and multipart/form-data formats - 根目录 > Image Generation > Unified interface > 302 Format V2 [Webhook Request Data Example](https://doc-en.302.ai/340677696e0.md): - 根目录 > Image Generation > Unified interface > 302 Format V2 [Asynchronous Image Generation Result Fetch](https://doc-en.302.ai/340677695e0.md): - 根目录 > Image Generation > Unified interface > Openai Format [Image Generation (OpenAI Format)](https://doc-en.302.ai/323639967e0.md): Model List & Parameter Instructions: - 根目录 > Image Generation > Unified interface > Openai Format [Image Editing (OpenAI Format)](https://doc-en.302.ai/326232197e0.md): Model List & Parameter Instructions: - 根目录 > Image Generation > GPT-Image Series [Generations（Image generation）](https://doc-en.302.ai/290106862e0.md): ### 1. Function Introduction - 根目录 > Image Generation > GPT-Image Series [Edits（Modify Image）](https://doc-en.302.ai/290106863e0.md): ### 1. Function Introduction - 根目录 > Image Generation > GPT-Image Series [Asynchronously fetch images](https://doc-en.302.ai/358653395e0.md): If `async=true` is passed in the generation API, use this API to retrieve the image. - 根目录 > Image Generation > DALL.E [Generations（DALL·E 3 and DALL·E 2）](https://doc-en.302.ai/207705140e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Image Generation > DALL.E [Edits（DALL·E 2） ](https://doc-en.302.ai/207705142e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Image Generation > DALL.E [Variations（DALL·E 2） ](https://doc-en.302.ai/207705141e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (Original Format - Text-to-Image)](https://doc-en.302.ai/379971011e0.md): The Nano-Banana-Pro model from Google - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (Original Format - Image Editing)](https://doc-en.302.ai/379974607e0.md): The Nano-Banana-Pro model from Google - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (Chat format)](https://doc-en.302.ai/380281445e0.md): The latest image generation model from Google’s Gemini series. It supports both text-to-image and image-to-image generation; you only need to pass the image URL via `image_url`. - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (simultaneous pay-per-use - text-to-image) Copy](https://doc-en.302.ai/381244057e0.md): The Nano-Banana-Pro model from Google - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (simultaneous pay-per-use - image editing) Copy](https://doc-en.302.ai/381244067e0.md): The Nano-Banana-Pro model from Google - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (asynchronous pay-per-use - text-to-image)](https://doc-en.302.ai/379974985e0.md): The Nano-Banana-Pro model from Google - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (asynchronous pay-per-use - image editing)](https://doc-en.302.ai/379975519e0.md): The Nano-Banana-Pro model from Google. - 根目录 > Image Generation > Google > Nano-Banana-Pro [gemini-3-pro-image-preview (asynchronous result retrieval)](https://doc-en.302.ai/379976951e0.md): In the request for the generation task, obtain the requestId and fill it into the API to get it. - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image (Original Format - Text-to-Image)](https://doc-en.302.ai/341704772e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image(Original Format - Image Editing)](https://doc-en.302.ai/341704773e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image (Pay-per-Use - Text-to-Image)](https://doc-en.302.ai/341704774e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image-edit (Pay-per-Use - Image Editing)](https://doc-en.302.ai/341704775e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image (asynchronous per-request billing - text-to-image)](https://doc-en.302.ai/342097222e0.md): From Google: gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image-edit (asynchronous per-request billing - image editing)](https://doc-en.302.ai/342096467e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Nano-Banana [gemini-2.5-flash-image (asynchronous result retrieval)](https://doc-en.302.ai/342096468e0.md): From Google's gemini-2.5-flash-image model - 根目录 > Image Generation > Google > Imagen [Imagen-4-Preview-Ultra (Image Generation)](https://doc-en.302.ai/341704776e0.md): From Google: Imagen-4-Preview-Ultra Model - 根目录 > Image Generation > Google > Imagen [Imagen-4-Preview-Fast (Image Generation)](https://doc-en.302.ai/341704777e0.md): From Google: Imagen-4-Preview-Fast model - 根目录 > Image Generation > Google > Imagen [Imagen-4-Preview （Image generated）](https://doc-en.302.ai/299579646e0.md): Imagen-4-Preview from Google - 根目录 > Image Generation > Google > Imagen [Imagen-3 (Image generated)](https://doc-en.302.ai/259641438e0.md): Imagen-3 model from Google - 根目录 > Image Generation > Google > Imagen [Imagen-3-Fast (Image generated)](https://doc-en.302.ai/259641439e0.md): Imagen-3-Fast model from Google - 根目录 > Image Generation > Stability.ai [Text-to-image (Image Generation-V1)](https://doc-en.302.ai/207705147e0.md): **Image creation using AI** - 根目录 > Image Generation > Stability.ai [Generate (Image Generation-SD2)](https://doc-en.302.ai/207705143e0.md): **Image creation using AI** - 根目录 > Image Generation > Stability.ai [Generate (Image Generation-SD3-Ultra)](https://doc-en.302.ai/207705146e0.md): **Image creation using AI V3.1, using model sd3** - 根目录 > Image Generation > Stability.ai [Generate (Image Generation-SD3)](https://doc-en.302.ai/225945514e0.md): Image generation through AI V3, using model SD3 - 根目录 > Image Generation > Stability.ai [Generate（Image Generation-SD3.5-Large）](https://doc-en.302.ai/207705144e0.md): Image generation through AI V3.5, using model SD3.5 - 根目录 > Image Generation > Stability.ai [Generate（Image Generation-SD3.5-Medium）](https://doc-en.302.ai/230033800e0.md): Image generation through AI V3.5, using model SD3.5 - 根目录 > Image Generation > Stability.ai [Generate（Image to Image-SD3） ](https://doc-en.302.ai/225948593e0.md): Image generation through AI V3, using model SD3 - 根目录 > Image Generation > Stability.ai [Generate（Image to Image-SD3.5-Large） ](https://doc-en.302.ai/207705145e0.md): Image generation through AI V3.5, using model SD3.5 - 根目录 > Image Generation > Stability.ai [Generate（Image to Image-SD3.5-Medium）](https://doc-en.302.ai/230033801e0.md): Image generation through AI V3.5, using model SD3.5 - 根目录 > Image Generation > Midjourney [Imagine](https://doc-en.302.ai/207705151e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney [Action](https://doc-en.302.ai/207705148e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney [Blend](https://doc-en.302.ai/207705149e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney [Describe](https://doc-en.302.ai/207705150e0.md): **Price: 0.025 PTC / call** - 根目录 > Image Generation > Midjourney [Modal](https://doc-en.302.ai/207705152e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney [Fetch](https://doc-en.302.ai/207705153e0.md): **Price: 0 PTC / call** - 根目录 > Image Generation > Midjourney [Cancel](https://doc-en.302.ai/207705154e0.md): **Price: 0 PTC / call** - 根目录 > Image Generation > Midjourney-Relax [Imagine](https://doc-en.302.ai/235831759e0.md): **Price：0.02PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Action](https://doc-en.302.ai/235831760e0.md): **Price：0.02PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Blend](https://doc-en.302.ai/235831761e0.md): **Price：0.02PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Describe](https://doc-en.302.ai/235831762e0.md): **Price：0.01PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Modal](https://doc-en.302.ai/235831763e0.md): **Price：0.02PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Fetch](https://doc-en.302.ai/235831764e0.md): **Price：0 PTC/call** - 根目录 > Image Generation > Midjourney-Relax [Cancel](https://doc-en.302.ai/235831765e0.md): **Price：0 PTC/call** - 根目录 > Image Generation > Midjourney-Turbo [Imagine](https://doc-en.302.ai/336195171e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Action](https://doc-en.302.ai/336195172e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Blend](https://doc-en.302.ai/336195174e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Describe](https://doc-en.302.ai/336195175e0.md): **Price: 0.025 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Modal](https://doc-en.302.ai/336195176e0.md): **Price: 0.05 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Fetch](https://doc-en.302.ai/336195177e0.md): **Price: 0 PTC / call** - 根目录 > Image Generation > Midjourney-Turbo [Cancel](https://doc-en.302.ai/336195178e0.md): **Price: 0 PTC / call** - 根目录 > Image Generation > 302.AI [Z-Image-Turbo (Image Generation - Tongyi)](https://doc-en.302.ai/383703844e0.md): The latest image generation model released by Tongyi Laboratory - 根目录 > Image Generation > 302.AI [SDXL](https://doc-en.302.ai/207705155e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SDXL-Lora](https://doc-en.302.ai/207705158e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SDXL-Lightning](https://doc-en.302.ai/207705161e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SDXL-Lightning-V2](https://doc-en.302.ai/207705162e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SDXL-Lightning-V3](https://doc-en.302.ai/207705163e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SD3](https://doc-en.302.ai/207705159e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SD3-V2](https://doc-en.302.ai/207705160e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Aura-Flow](https://doc-en.302.ai/207705156e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Kolors](https://doc-en.302.ai/207705157e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Kolors（Reference Image Generation-KLING）](https://doc-en.302.ai/247559341e0.md): 02.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [QRCode Generation](https://doc-en.302.ai/212293001e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Lora](https://doc-en.302.ai/224673634e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Lora（Get task result）](https://doc-en.302.ai/335020753e0.md): **Price: 0 PTC/time** - 根目录 > Image Generation > 302.AI [SD-3.5-Large](https://doc-en.302.ai/230035306e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SD-3.5-Large-Turbo](https://doc-en.302.ai/230035307e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [SD-3.5-Medium](https://doc-en.302.ai/230035308e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > 302.AI [Lumina-Image-V2(Image generated)](https://doc-en.302.ai/259285282e0.md): 302.AI's API, comes from the model we deploy ourselves on cloud GPUs. Some of the models are open-source, and some of them are fine-tuned or developed by ourselves. - 根目录 > Image Generation > 302.AI [Playground-v25（Image generated）](https://doc-en.302.ai/267233314e0.md): Open-source image generation model Playground - 根目录 > Image Generation > 302.AI [Omnigen-V1（Image generated）](https://doc-en.302.ai/267354725e0.md): Open-source image generation model Omnigen - 根目录 > Image Generation > 302.AI [Qwen-Image（image generation）](https://doc-en.302.ai/331356565e0.md): 302.AI's API is powered by models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Generation > 302.AI [Qwen-Image-2512 (Image Generation)](https://doc-en.302.ai/401658100e0.md): 302.AI's API comes from models we deployed ourselves on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Generation > 302.AI [Qwen-Image-Lora (LoRA image generation)](https://doc-en.302.ai/337808543e0.md): The API of 302.AI comes from models we have deployed on cloud GPUs. Some models are open source, while others have been fine-tuned or developed by us. - 根目录 > Image Generation > 302.AI [Qwen-Image-Lora-Trainer (LoRA training)](https://doc-en.302.ai/337808544e0.md): Used for image generation, allows training your own lora - 根目录 > Image Generation > 302.AI [Qwen-Image-Lora-Trainer (asynchronous result retrieval)](https://doc-en.302.ai/337808545e0.md): Get Lora Training Results - 根目录 > Image Generation > Glif [Glif（Claude+SD3）](https://doc-en.302.ai/207705168e0.md): Automatically optimize prompts through Claude, then use SD3 for drawing. - 根目录 > Image Generation > Glif [Glif (Text-to-Sticker)](https://doc-en.302.ai/207705164e0.md): **Input description, generate sticker image** - 根目录 > Image Generation > Glif [Glif (Text-to-Graffiti)](https://doc-en.302.ai/207705165e0.md): **Input description, generate doodle image** - 根目录 > Image Generation > Glif [Glif (Text-to-Wojak Comic)](https://doc-en.302.ai/207705166e0.md): **Input description, generate Wojak comic** - 根目录 > Image Generation > Glif [Glif (Text-to-Lego)](https://doc-en.302.ai/207705167e0.md): **Input description, generate Lego image** - 根目录 > Image Generation > Flux > Official API [Generate](https://doc-en.302.ai/256980713e0.md): Official documentation: https://api.bfl.ml/scalar#tag/tasks/POST/v1/flux-pro - 根目录 > Image Generation > Flux > Official API [Finetune](https://doc-en.302.ai/256980714e0.md): Official documentation: https://api.bfl.ml/scalar#tag/tasks/POST/v1/flux-pro - 根目录 > Image Generation > Flux > Official API [Result](https://doc-en.302.ai/256980715e0.md): **Price: Free** - 根目录 > Image Generation > Flux [Flux-2-Max (generates images)](https://doc-en.302.ai/394145328e0.md): Official Documentation: https://api.bfl.ml/scalar#tag/models/POST/v1/flux-2-max - 根目录 > Image Generation > Flux [Flux-2-Max (Query Task)](https://doc-en.302.ai/394145666e0.md): **Price: Free** - 根目录 > Image Generation > Flux [Flux-2-Flex (generates images)](https://doc-en.302.ai/383498869e0.md): Official documentation: https://api.bfl.ml/scalar#tag/tasks/POST/v1/flux-pro - 根目录 > Image Generation > Flux [Flux-2-Flex (Query Task)](https://doc-en.302.ai/383499541e0.md): **Price: Free** - 根目录 > Image Generation > Flux [Flux-2-Pro (generates images)](https://doc-en.302.ai/383499992e0.md): Official documentation: https://api.bfl.ml/scalar#tag/tasks/POST/v1/flux-pro - 根目录 > Image Generation > Flux [Flux-2-Pro (query task)](https://doc-en.302.ai/383500161e0.md): **Price: Free** - 根目录 > Image Generation > Flux [Flux-Ultra（v1.1）](https://doc-en.302.ai/231457732e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Pro（v1.1）](https://doc-en.302.ai/224673635e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Pro](https://doc-en.302.ai/207705169e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Dev](https://doc-en.302.ai/207705170e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Schnell](https://doc-en.302.ai/207705172e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Realism](https://doc-en.302.ai/207705171e0.md): Created by black forest labs, founded by former Stability.ai members, for image generation, from: https://blackforestlabs.ai/ - 根目录 > Image Generation > Flux [Flux-Lora](https://doc-en.302.ai/207705173e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > Flux [Flux-General](https://doc-en.302.ai/232136668e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > Flux [Flux-General-Inpainting（Advanced Customization）](https://doc-en.302.ai/251739170e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Generation > Flux [Flux-Lora-Training（Training Lora）](https://doc-en.302.ai/252473800e0.md): LoRA for training your own image generation - 根目录 > Image Generation > Flux [Flux-Lora-Training（Fetch Results Asynchronously）](https://doc-en.302.ai/252474289e0.md): Fetch Lora training result - 根目录 > Image Generation > Flux [Flux-1-Krea（Image Generation）](https://doc-en.302.ai/330039597e0.md): Krea1 is a model jointly developed by Krea AI and Black Forest Labs (FLUX series developers), aimed at addressing the "AI feel" issue commonly found in traditional AI image generation. - 根目录 > Image Generation > Flux [Flux-1-Krea-Redux（Image-to-Image Generation）](https://doc-en.302.ai/330131542e0.md): Krea1 is a model jointly developed by Krea AI and Black Forest Labs (FLUX series developers), aimed at addressing the "AI feel" issue commonly found in traditional AI image generation. - 根目录 > Image Generation > Flux [Flux-1-SRPO（Image-to-Image Generation）](https://doc-en.302.ai/352620717e0.md): Flux-1-SRPO, Tencent’s latest research, designed to address the “AI look” problem in traditional AI image generation - 根目录 > Image Generation > Ideogram [Generate（subject reference）](https://doc-en.302.ai/337880895e0.md): Text-to-image generation from Ideogram excels at producing accurate text and posters. It now supports inputting portrait subjects to generate images referencing the subject. - 根目录 > Image Generation > Ideogram [Generate（Text-to-image V3）](https://doc-en.302.ai/292492001e0.md): Text-to-image generation from Ideogram excels at producing accurate text and posters. - 根目录 > Image Generation > Ideogram [Generate（Text-to-image）](https://doc-en.302.ai/212283108e0.md): Text-to-image generation from Ideogram, with the advantage of producing accurate text and posters - 根目录 > Image Generation > Recraft [Recraft-V3（Text to Image）](https://doc-en.302.ai/229153719e0.md): The latest enigmatic model raising eyebrows on the LLM big model list is "red_panda."，project from: https://www.recraft.ai/ - 根目录 > Image Generation > Recraft [Create-Style（Customized Styles）](https://doc-en.302.ai/229153720e0.md): The latest enigmatic model raising eyebrows on the LLM big model list is "red_panda."，project from: https://www.recraft.ai/ - 根目录 > Image Generation > Recraft [Recraft-20B（Image Generation）](https://doc-en.302.ai/245633638e0.md): The new version of Recraft is 40% cheaper than the original version, but it's not as effective as the original - 根目录 > Image Generation > Luma [Luma-Photon（Image generation）](https://doc-en.302.ai/240589098e0.md): Image generation model from Luma - 根目录 > Image Generation > Luma [Luma-Photon-Flash（Fast image generation）](https://doc-en.302.ai/240589099e0.md): Image generation model from Luma - 根目录 > Image Generation > Doubao [Generations（Seedream 4.5）](https://doc-en.302.ai/385938976e0.md): The latest model from Doubao Seedream supports both text-generated and image-generated images. - 根目录 > Image Generation > Doubao [Generations（Seedream 4.0）](https://doc-en.302.ai/347888792e0.md): YouDream Latest Model Supporting Text-to-Image and Image-to-Image - 根目录 > Image Generation > Doubao [Generations（Seedream 3.0）](https://doc-en.302.ai/318394376e0.md): The newly released JiMeng API adopts an OpenAI-compatible image generation format and offers higher concurrency. - 根目录 > Image Generation > Doubao [Generations（SeedEdit 3.0）](https://doc-en.302.ai/334800559e0.md): The newly released API format by Jimeng adopts an OpenAI-compatible image generation format, with higher concurrency. - 根目录 > Image Generation > Doubao [Drawing（Doubao image generation） ](https://doc-en.302.ai/246609554e0.md): Image generation model from Doubao - 根目录 > Image Generation > Minimax [image（Text-to-Image Generation）](https://doc-en.302.ai/270134792e0.md): A text-to-image model from Minimax, supported models: - 根目录 > Image Generation > ZHIPU [image（Text-to-Image Generation）](https://doc-en.302.ai/270884894e0.md): Image generation model from CogView-4, supported models: - 根目录 > Image Generation > Baidu [iRAG(Text-to-Image Generation)](https://doc-en.302.ai/279708373e0.md): This feature uses Baidu's image generation model iRAG, which retrieves real images before generating new ones. This helps reduce hallucinations and improves realism. - 根目录 > Image Generation > Hidream [Hidream-i1-full（Advanced Version）](https://doc-en.302.ai/283664126e0.md): Image Generation Model from Zhixiang Future - 根目录 > Image Generation > Hidream [Hidream-i1-dev（Intermediate Version）](https://doc-en.302.ai/283664440e0.md): Image Generation Model from Zhixiang Future - 根目录 > Image Generation > Hidream [Hidream-i1-fast（Entry-Level Version）](https://doc-en.302.ai/283664480e0.md): Image Generation Model from Zhixiang Future - 根目录 > Image Generation > Bagel [Bagel（Image generation）](https://doc-en.302.ai/300878512e0.md): Image model from Byte Open Source, similar to gpt-image-1, can generate images and modify them - 根目录 > Image Generation > SiliconFlow [Create Image Generation Request ](https://doc-en.302.ai/308965989e0.md): Model: Kwai-Kolors/Kolors - 根目录 > Image Generation > Higgsfield > Official Format [Soul (Text-to-Image)](https://doc-en.302.ai/393934724e0.md): Image source: higgsfield.ai - 根目录 > Image Generation > Higgsfield > Official Format [Fetch (Get task results)](https://doc-en.302.ai/393934725e0.md): Obtain the results of image generation or feature generation tasks - 根目录 > Image Generation > Higgsfield > Official Format [Styles (Get Style List)](https://doc-en.302.ai/393934726e0.md): **Price: Free / call** - 根目录 > Image Generation > Higgsfield [Soul (Text to Image)](https://doc-en.302.ai/316337394e0.md): Text-to-image from higgsfield.ai - 根目录 > Image Generation > Higgsfield [Character (Generate Character)](https://doc-en.302.ai/323639968e0.md): Character generation from higgsfield.ai supports inputting up to 70 face images. The AI will analyze your images and, within **a few minutes**, understand your unique appearance to generate a personalized image model. - 根目录 > Image Generation > Higgsfield [Apps (Image-to-Image)](https://doc-en.302.ai/347477579e0.md): Automatically edit images based on predefined templates - 根目录 > Image Generation > Higgsfield [Fetch (Retrieve task results)](https://doc-en.302.ai/316337395e0.md): Fetch video generation task results - 根目录 > Image Generation > Kling > 302 format [Images-Generations（o1-Images-Generations）](https://doc-en.302.ai/386831371e0.md): From Kling's Kling o1, you can upload 1-10 reference images and use natural language commands to add, delete, modify, merge, and stylize them, generating a completely new image with a single click. You can also directly input text commands, and the model will perfectly depict the image you need. - 根目录 > Image Generation > Kling > 302 format [Fetch（Get task results）](https://doc-en.302.ai/386831372e0.md): Video from Kling - 根目录 > Image Generation > Kling [Images-Generations (Image Generation)](https://doc-en.302.ai/322532993e0.md): # Kling Image Generation - 根目录 > Image Generation > Kling [Fetch (Get Generated Image Task Result)](https://doc-en.302.ai/322532994e0.md): From Kling Raw Images - 根目录 > Image Generation > Tongyi Wanxiang [Qwen-Image-Max (text-to-image)](https://doc-en.302.ai/400547056e0.md): Qwen-Image is a versatile text-to-image generation model that supports multiple artistic styles, with particular strengths in complex text rendering. The model enables multi-line layouts, paragraph-level text generation, and fine-grained detail depiction, making it capable of realizing sophisticated text-image hybrid layout designs. - 根目录 > Image Generation > Tongyi Wanxiang [Qwen-Image (Alibaba Cloud Deployment)](https://doc-en.302.ai/340677697e0.md): Qwen-Image is a general-purpose image generation model that supports a variety of artistic styles and excels at complex text rendering, especially for both Chinese and English text. The model supports multi-line layouts, paragraph-level text generation, and fine-grained detail depiction, enabling complex mixed text-image layout designs. - 根目录 > Image Generation > Tongyi Wanxiang [T2I (Text-to-Image Alibaba Cloud)](https://doc-en.302.ai/333236477e0.md): Tongyi Wanxiang - Text-to-Image model generates images based on text. It is recommended to choose the newly upgraded Wanxiang 2.2 model to start your AI journey in text-to-image generation. - 根目录 > Image Generation > Tongyi Wanxiang [Wan2.6-Image (Image Generation and Editing)](https://doc-en.302.ai/393705688e0.md): Tongyi Wanxiang image generation model supports image editing and mixed text-image output, meeting diverse generation and integration needs. - 根目录 > Image Generation > Tongyi Wanxiang [Tasks (Retrieve Task Results)](https://doc-en.302.ai/333236478e0.md): Get Task Result - 根目录 > Image Generation > Vidu [Reference to Image (Reference2Image)](https://doc-en.302.ai/350722876e0.md): Refer to the official documentation: https://platform.vidu.cn/docs/reference-to-image - 根目录 > Image Generation > Vidu [Fetch V2 (Fetch task results)](https://doc-en.302.ai/350722877e0.md): Get Video Generation Task Result - 根目录 > Image Generation > Wavespeed > Universal Interface [Wavespeed Generation Task](https://doc-en.302.ai/359497922e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Image Generation > Wavespeed > Universal Interface [Wavespeed Retrieval Task](https://doc-en.302.ai/359497923e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Image Generation > Wavespeed [alibaba/wan-2.5/text-to-image](https://doc-en.302.ai/363568411e0.md): Generate AI images using Alibaba WAN 2.5 text-to-image model. - 根目录 > Image Generation > Wavespeed [bytedance/seedream-v4/sequential](https://doc-en.302.ai/364076801e0.md): Seedream 4.0, supports 4K resolution image generation · editing · character and object consistency · multi-image sequence generation. - 根目录 > Image Generation > Wavespeed [bytedance/dreamina-v3.0/text-to-image](https://doc-en.302.ai/364888142e0.md): ByteDance Dreamina V3.0 Text-to-Image is a text-to-image model based on the same technology as Dreamina, focusing on enhancing the visual quality of generated images. This version features significant improvements in image aesthetics, style accuracy, and detail richness, while maintaining responsiveness to text input. - 根目录 > Image Generation > Wavespeed [bytedance/dreamina-v3.1/text-to-image](https://doc-en.302.ai/364888141e0.md): ByteDance Dreamina V3.1 Text-to-Image is a text-to-image model based on the same technology as Dreamina, focusing on enhancing the visual quality of generated images. This version features significant improvements in image aesthetics, style accuracy, and detail richness, while maintaining responsiveness to text input. - 根目录 > Image Generation > Wavespeed [bytedance/seedream-v3.1](https://doc-en.302.ai/365653478e0.md): ByteDance Seedream V3.1 is a text-to-image model derived from the same technology as Seedream, focusing on upgrading the visual effects of generated images. This version has made significant improvements in image aesthetics, style accuracy, and detail richness, while maintaining responsiveness to text input. - 根目录 > Image Generation > Wavespeed [bytedance/waver-1.0](https://doc-en.302.ai/365653480e0.md): Waver1.0 is ByteDance's versatile video generation model, specializing in text-to-video (T2V), image-to-video (I2V), and text-to-image (T2I) generation. - 根目录 > Image Generation > Wavespeed [bytedance/seedream-v4](https://doc-en.302.ai/363568410e0.md): Seedream 4.0 is the most advanced image model launched by ByteDance. Seedream 4.0: surpasses Nano Banana in every aspect. - 根目录 > Image Generation > Wavespeed [google/gemini-2.5-flash-image-preview/text-to-image](https://doc-en.302.ai/364419462e0.md): Google's state-of-the-art image generation and editing model. - 根目录 > Image Generation > Wavespeed [google/imagen4](https://doc-en.302.ai/364076802e0.md): Google's Imagen 4 flagship model - 根目录 > Image Generation > Wavespeed [google/nano-banana/text-to-image](https://doc-en.302.ai/363568412e0.md): Google's Nano Banana is a cutting-edge text-to-image generation model. - 根目录 > Image Generation > Wavespeed [google/imagen4-fast](https://doc-en.302.ai/367312081e0.md): Google's highest quality image generation model - 根目录 > Image Generation > Wavespeed [google/imagen4-ultra](https://doc-en.302.ai/367312080e0.md): Google's highest quality image generation model - 根目录 > Image Generation > Wavespeed [google/imagen3-fast](https://doc-en.302.ai/367312083e0.md): Google's highest-quality text-to-image model, capable of generating images with rich details, outstanding lighting effects, and visually appealing results - 根目录 > Image Generation > Wavespeed [google/gemini-2.5-flash-image/text-to-image](https://doc-en.302.ai/364888140e0.md): Google Gemini 2.5 Flash Image is a powerful new image generation and editing model, featuring advanced capabilities and creative control. - 根目录 > Image Generation > Wavespeed [google/imagen3](https://doc-en.302.ai/367312084e0.md): Google's highest-quality text-to-image model, capable of generating images with rich details, outstanding lighting effects, and visually appealing results - 根目录 > Image Generation > Wavespeed [ideogram-ai/ideogram-v2a-turbo](https://doc-en.302.ai/367223202e0.md): The highest quality Ideogram v3 model. - 根目录 > Image Generation > Wavespeed [ideogram-ai/ideogram-v3-turbo](https://doc-en.302.ai/367223199e0.md): The highest quality Ideogram v3 model. - 根目录 > Image Generation > Wavespeed [ideogram-ai/ideogram-v3-balanced](https://doc-en.302.ai/367223198e0.md): The highest quality Ideogram v3 model. - 根目录 > Image Generation > Wavespeed [ideogram-ai/ideogram-v2](https://doc-en.302.ai/367223203e0.md): An excellent image model featuring state-of-the-art image restoration, prompt understanding, and text rendering technologies. - 根目录 > Image Generation > Wavespeed [ideogram-ai/ideogram-v3-quality](https://doc-en.302.ai/367223200e0.md): The highest quality Ideogram v3 model. - 根目录 > Image Generation > Wavespeed [leonardoai/phoenix-1.0](https://doc-en.302.ai/364217870e0.md): Phoenix offers excellent prompt adherence, coherent text within images, and provides iterative prompt capabilities through AI editing. Phoenix is Leonardo. - 根目录 > Image Generation > Wavespeed [leonardoai/lucid-origin](https://doc-en.302.ai/364217871e0.md): Experience visually stunning artistic creations with exceptional quality, outstanding prompt adherence accuracy, broader creative diversity, and impressive detail and definition. - 根目录 > Image Generation > Wavespeed [luma/photon-flash](https://doc-en.302.ai/367312085e0.md): Luma Photon Flash text-to-image model - 根目录 > Image Generation > Wavespeed [luma/photon](https://doc-en.302.ai/367312086e0.md): Luma Photon Text-to-Image Model - 根目录 > Image Generation > Wavespeed [openai/dall-e-2](https://doc-en.302.ai/365653475e0.md): Original Classic DALL·E 2 - 根目录 > Image Generation > Wavespeed [openai/dall-e-3](https://doc-en.302.ai/365653476e0.md): Original Classic DALL·E 3 - 根目录 > Image Generation > Wavespeed [openai/gpt-image-1/text-to-image](https://doc-en.302.ai/365653477e0.md): OpenAI's latest image generation model: gpt-1-image - 根目录 > Image Generation > Wavespeed [openai/gpt-image-1-high-fidelity](https://doc-en.302.ai/364419463e0.md): Google's state-of-the-art image generation and editing model. - 根目录 > Image Generation > Wavespeed [reve/text-to-image](https://doc-en.302.ai/364217869e0.md): Reve Art AI - Generate beautiful images using Reve Art AI. - 根目录 > Image Generation > Wavespeed [recraft-ai/recraft-v3](https://doc-en.302.ai/367312078e0.md): Recraft V3 (codenamed red_panda) is a text-to-image model capable of generating long texts and images in various styles. As of now, it is at the SOTA level in the field of image generation, as demonstrated by the text-to-image benchmark tests from Artificial Analysis. - 根目录 > Image Generation > Wavespeed [recraft-ai/recraft-v3-svg](https://doc-en.302.ai/367312079e0.md): Recraft V3 SVG is a text-to-image model capable of generating long text and images in various styles. As of now, it is at the SOTA level in the field of image generation, as confirmed by the text-to-image benchmark from Artificial Analysis. - 根目录 > Image Generation > Wavespeed [recraft-ai/recraft-20b-svg](https://doc-en.302.ai/367223204e0.md): Affordable and fast vector images - 根目录 > Image Generation > Wavespeed [runwayml/gen4-image](https://doc-en.302.ai/365103925e0.md): Runway's Gen-4 image model supports reference image functionality. You can use up to 3 reference images to create the precise image you need, capturing every angle. - 根目录 > Image Generation > Wavespeed [recraft-ai/recraft-20b](https://doc-en.302.ai/367312077e0.md): Affordable and fast images - 根目录 > Image Generation > Wavespeed [stability-ai/stable-diffusion-3.5-large-turbo](https://doc-en.302.ai/365103927e0.md): A text-to-image model capable of generating high-resolution, detail-rich images. It supports multiple artistic styles and can produce diverse outputs from the same prompt, while focusing on reducing inference steps. - 根目录 > Image Generation > Wavespeed [stability-ai/stable-diffusion-3.5-medium](https://doc-en.302.ai/365103926e0.md): An image model with 2.5 billion parameters, utilizing an improved MMDiT-X architecture. - 根目录 > Image Generation > Wavespeed [stability-ai/stable-diffusion-3.5-large](https://doc-en.302.ai/365151860e0.md): 一款文本到图像模型，能够生成高分辨率、细节精致的图像。它支持多种艺术风格，并能从相同的提示词生成多样化的输出，这得益于 Query-Key Normalization 技术。 - 根目录 > Image Generation > Wavespeed [stability-ai/stable-diffusion-3](https://doc-en.302.ai/365151861e0.md): 一款文本到图像模型，能够生成高分辨率、细节精致的图像。它支持多种艺术风格，并能从相同的提示词生成多样化的输出，这得益于 Query-Key Normalization 技术。 - 根目录 > Image Generation > Wavespeed [stability-ai/stable-diffusion](https://doc-en.302.ai/365151862e0.md): 一款潜在文本到图像扩散模型，能够根据任何文本输入生成照片级真实感的图像。 - 根目录 > Image Generation > Wavespeed [stability-ai/sdxl](https://doc-en.302.ai/365151863e0.md): 一款潜在文本到图像扩散模型，能够根据任何文本输入生成照片级真实感的图像。 - 根目录 > Image Generation > Wavespeed [stability-ai/sdxl-lora](https://doc-en.302.ai/364888143e0.md): An AI model for text-to-image generation that supports LoRA, capable of creating beautiful images. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/any-llm/vision](https://doc-en.302.ai/364419461e0.md): Use any visual language model from our curated catalog (powered by OpenRouter). - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-dev-lora](https://doc-en.302.ai/363698931e0.md): FLUX.1 [dev] Ultra-fast model endpoint, supports LoRA, capable of utilizing pre-trained LoRA adapters for personalized, high-quality image generation, suitable for specific styles, brand identity, and product-specific outputs. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-dev-ultra-fast](https://doc-en.302.ai/360850249e0.md): FLUX.1 [dev] is a Rectified Flow Transformer model with 12 billion parameters, capable of generating high-quality images based on text descriptions. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/qwen-image/text-to-image](https://doc-en.302.ai/363698932e0.md): Qwen-Image is a 20B MMDiT model designed for next-generation text-to-image generation. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/wan-2.2/text-to-image-lora](https://doc-en.302.ai/363698933e0.md): Generate highly detailed images using WAN 2.2, with support for custom LoRAs. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/hunyuan-image-3](https://doc-en.302.ai/363698934e0.md): HunyuanImage-3.0 is a groundbreaking native multimodal model that unifies multimodal understanding and generation within an autoregressive framework. Our text-to-image module matches or surpasses the performance of leading closed-source models. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/wan-2.2/text-to-image-realism](https://doc-en.302.ai/363698935e0.md): Generate surreal images from text prompts using WAN 2.2. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-dev](https://doc-en.302.ai/364076799e0.md): FLUX.1 [dev] is a correction flow transformer with 12 billion parameters, capable of generating images based on text descriptions. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/qwen-image/text-to-image-lora](https://doc-en.302.ai/364076800e0.md): Qwen-Image LoRa is a 20B MMDiT model that combines LoRA technology for next-generation text-to-image generation. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-dev-lora-ultra-fast](https://doc-en.302.ai/364076803e0.md): FLUX.1 [dev] Ultra-fast model endpoint supporting LoRA, utilizing pre-trained LoRA adapters for personalized, high-quality image generation, suitable for specific styles, brand identity, and product-specific outputs. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-1-srpo](https://doc-en.302.ai/364217872e0.md): Experience visually stunning artistic creations with exceptional quality, outstanding prompt adherence accuracy, broader creative diversity, and impressive detail and definition. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-srpo](https://doc-en.302.ai/364419458e0.md): FLUX SRPO [dev] is a state-of-the-art 12 billion parameter flow transformer designed to generate stunning, high-quality images from text, featuring exceptional aesthetic performance. Perfect for personal and commercial use. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/female-human](https://doc-en.302.ai/364419459e0.md): Advanced female human character generation model for creating realistic and diverse female portraits, featuring high quality and fine details. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/hunyuan-image-2.1](https://doc-en.302.ai/364419460e0.md): HunyuanImage-2.1: An efficient diffusion model for high-resolution (2K) text-to-image generation. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/chroma](https://doc-en.302.ai/364888139e0.md): The Chroma model supports completely unrestricted image generation, enabling creators to freely express their imagination and artistic creativity. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/qwen-image/edit-plus-lora](https://doc-en.302.ai/365653479e0.md): Qwen-Image-Edit-Plus is a 20B MMDiT model designed for next-generation image editing and generation, featuring improved multi-image editing, single-image consistency, and native support for ControlNet. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/wan-2.2/i2v-5b-720p](https://doc-en.302.ai/365653481e0.md): Generate unlimited AI videos using the Wan 2.2 image-to-video model. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-krea-dev-lora](https://doc-en.302.ai/366328038e0.md): FLUX.1 Krea [dev] is the open-source weights version of Krea 1, featuring powerful performance, a highly recognizable aesthetic style, and outstanding realism. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-1.1-pro-ultra](https://doc-en.302.ai/366328039e0.md): FLUX 1.1 Pro Ultra is the latest version in the FLUX series, offering professional-grade image quality, enhanced photo-realistic effects, and support for up to 2K resolution. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-1.1-pro](https://doc-en.302.ai/366328040e0.md): FLUX 1.1 Pro is the latest version in the FLUX series, offering professional-grade image quality with enhanced photo-realistic rendering and support for up to 2K resolution. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-pro/text-to-image](https://doc-en.302.ai/382557075e0.md): Black Forest Labs' FLUX.2 [pro] offers production - grade text - to - image generation, with enhanced realism, clearer text rendering, and native editing, achieving reliable and reproducible results. It features production - level text - to - image capabilities, studio - quality output, and zero parameter hassle. After calibration, only a good prompt is needed — no need for guessing guidance scales or experimenting with steps — just reliable and event - suitable images. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-flex/text-to-image](https://doc-en.302.ai/382557334e0.md): Black Forest Labs' FLUX.2 [flex] offers fast and flexible text-to-image generation, enhanced realism, clearer text rendering, and built-in editing support for rapid iteration: ready-to-use REST inference API, optimal performance, no cold start, and affordable pricing. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-dev/text-to-image](https://doc-en.302.ai/382557997e0.md): FLUX.2 [dev] is a streamlined base model in the FLUX.2 family: an open-source text-to-image engine, fine-tuned with a focus on speed, stability, and training friendliness. It delivers solid visual quality while remaining compact enough for rapid iteration, LoRA experiments, and large-scale batch processing. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-pro/edit](https://doc-en.302.ai/382576870e0.md): FLUX.2 [pro] Edit is the top-tier image editing engine of the FLUX family, built for production teams who need reliable, high-fidelity edits from natural language instructions. It can take one or several reference images plus a prompt and turn them into clean, on-brand transformations suitable for final delivery—not just drafts. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-flex/edit](https://doc-en.302.ai/382576871e0.md): FLUX.2 [flex] Edit is a configurable image editing model built on FLUX.2 [flex], aimed at teams that need fine control over how edits are applied. It can work with one or multiple reference images and lets you tune quality–speed trade-offs, making it a good fit for style-heavy, budget-conscious production workflows. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-2-dev/edit](https://doc-en.302.ai/382576872e0.md): FLUX.2 [dev] Edit is the lean editing companion to FLUX.2 [dev] Text-to-Image: a lightweight, open-source model that updates existing images quickly while keeping their core look intact. It is built for day-to-day production tasks where you need reliable edits at scale without burning GPU budget. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/neta-lumina](https://doc-en.302.ai/366328041e0.md): Generate images using Neta Lumina—a high-quality anime-style image generation model developed by Neta.art Lab. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/wan-2.1/text-to-image-lora](https://doc-en.302.ai/367328673e0.md): Revolutionary text-to-image generation technology powered by Wan 2.1, capable of producing ultra-realistic images with photographic authenticity and exceptional detail fidelity. We have modified the Wan 2.1 video model to also support image generation, and found that it can achieve SOTA image generation quality! This endpoint also supports LoRAs. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/wan-2.1/text-to-image](https://doc-en.302.ai/367328674e0.md): Revolutionary text-to-image generation technology powered by Wan 2.1, capable of producing ultra-realistic images with photographic authenticity and exceptional detail fidelity. We have modified the Wan 2.1 video model to also support image generation, and found that it achieves SOTA image generation quality! - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-schnell](https://doc-en.302.ai/367761203e0.md): FLUX.1 [schnell] is a rectified flow transformer with 12 billion parameters that can generate images based on text descriptions. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-kontext-max/text-to-image](https://doc-en.302.ai/367761204e0.md): FLUX.1 Kontext [max] text-to-image is a brand-new advanced model that delivers exceptional performance in every aspect—significantly improving adherence to prompts. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/flux-kontext-pro/text-to-image](https://doc-en.302.ai/367761205e0.md): FLUX.1 Kontext [pro] Text-to-Image is a brand-new advanced model that delivers ultimate performance in every aspect—significantly improving adherence to prompts. - 根目录 > Image Generation > Wavespeed [bytedance/seedream-v3](https://doc-en.302.ai/367761202e0.md): ByteDance Seedream V3 is an advanced text-to-image model, excelling at generating high-quality, realistic images with exceptional detail and artistic style. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/imagen4](https://doc-en.302.ai/367761206e0.md): Imagen 4 is an AI image generator capable of creating detailed, high-quality images in various styles with resolutions up to 2K. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/step1x-edit](https://doc-en.302.ai/367761207e0.md): Step1X-Edit can edit your photos based on simple instructions, transforming them into stunning, professional-grade works that rival top proprietary tools. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/hidream-i1-dev](https://doc-en.302.ai/367761208e0.md): HiDream-I1 is a new open-source image generation foundation model with 17 billion parameters, capable of delivering state-of-the-art image generation quality within seconds. - 根目录 > Image Generation > Wavespeed [wavespeed-ai/hidream-i1-full](https://doc-en.302.ai/367761209e0.md): HiDream-I1 is a new open-source image generation foundation model with 17 billion parameters, capable of achieving state-of-the-art image generation quality within seconds. - 根目录 > Image Generation > Wavespeed [Wavespeed Retrieval Task](https://doc-en.302.ai/364223439e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Outfit Change Task](https://doc-en.302.ai/270667213e0.md): Outfit-changing effect achieved through a complex ComfyUI workflow. Commercial-grade quality, suitable for model outfit changes. Execution time: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Outfit Change Task (Upload Mask)](https://doc-en.302.ai/282245359e0.md): Achieve outfit change effect through a complex workflow in ComfyUI, providing commercial-grade effects suitable for model outfit changes. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Query Outfit Change Task Status](https://doc-en.302.ai/270354432e0.md): Outfit-changing effect achieved through a complex ComfyUI workflow. Commercial-grade quality, suitable for model outfit changes. Execution time: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Face Swap Task](https://doc-en.302.ai/270356436e0.md): Face swap effect achieved through complex ComfyUI workflow, commercial-grade quality, suitable for model face swapping, runtime 3-5 minutes - 根目录 > Image Processing > 302.AI-ComfyUI [Query Face Swap Task Status](https://doc-en.302.ai/270356569e0.md): Face swap effect achieved through a complex ComfyUI workflow, with commercial-grade quality, suitable for model face swapping. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create a Task to Replace Any Item](https://doc-en.302.ai/270356644e0.md): Achieve object replacement effects through a complex ComfyUI workflow, delivering commercial-grade results suitable for still-life advertisements. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Object Replacement Task (Upload Mask)](https://doc-en.302.ai/282245422e0.md): Achieve object replacement effect through a complex workflow in ComfyUI, providing commercial-grade effects suitable for still life advertisements and other scenarios. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Check the Status of Any Object Replacement Task](https://doc-en.302.ai/270356665e0.md): The object replacement effect is achieved through a complex ComfyUI workflow, providing commercial-grade results suitable for still-life advertisements and similar scenarios. The runtime is approximately 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create a Task to Transform Cartoon Characters into Real People](https://doc-en.302.ai/272720420e0.md): A commercial-grade effect achieved through a complex ComfyUI workflow that transforms cartoons into realistic human images. Can convert clothing design drawings into images with real models. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Query the status of the task to turn a manga character into a real person](https://doc-en.302.ai/272722166e0.md): The manga-to-real-person effect is achieved through a complex workflow in ComfyUI, delivering commercial-grade results. It can convert clothing design sketches into real model images, with a runtime of 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Style Transfer Task](https://doc-en.302.ai/272722220e0.md): Image style transfer achieved through a complex workflow with ComfyUI, delivering commercial-grade results. Runtime: 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Query the status of the style transfer task](https://doc-en.302.ai/272722249e0.md): Image style transfer implemented through a complex ComfyUI workflow, achieving commercial-grade results with a runtime of 3-5 minutes. - 根目录 > Image Processing > 302.AI-ComfyUI [Create Image Removal Task](https://doc-en.302.ai/289966914e0.md): **Price: 0.01 PTC/call** - 根目录 > Image Processing > 302.AI-ComfyUI [Query Image Removal Task Status](https://doc-en.302.ai/289966965e0.md): **Price：free** - 根目录 > Image Processing > 302.AI [Qwen-Image-Layered](https://doc-en.302.ai/394728849e0.md): 302.AI's API comes from models we deployed ourselves on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Qwen-Image-Layered（Image Layering Results）](https://doc-en.302.ai/394728850e0.md): **Price: 0 PTC/Time** - 根目录 > Image Processing > 302.AI [Qwen-Image-Edit](https://doc-en.302.ai/337808542e0.md): 302.AI's API comes from models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Qwen-Image-Edit-Plus (Image Editing)](https://doc-en.302.ai/354583211e0.md): The API of 302.AI comes from models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by ourselves. - 根目录 > Image Processing > 302.AI [Upscale](https://doc-en.302.ai/207705177e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Upscale-V2](https://doc-en.302.ai/207705178e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Upscale-V3](https://doc-en.302.ai/207705179e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Upscale-V4](https://doc-en.302.ai/207705180e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Upscale-V5](https://doc-en.302.ai/359336021e0.md): 302.AI's API comes from models we have deployed on cloud GPU. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Upscale-V6](https://doc-en.302.ai/373210515e0.md): The API of 302.AI comes from the models we deployed ourselves on cloud GPUs. Some of the models are open-source, and some are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Upscale-Fast (Fast Image Enlargement)](https://doc-en.302.ai/373456906e0.md): The 302.AI API comes from the models we deployed ourselves on cloud GPUs. Some of these models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Super-Upscale](https://doc-en.302.ai/207705181e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Super-Upscale-V2](https://doc-en.302.ai/207705182e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Face-upscale](https://doc-en.302.ai/207705185e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Trellis（Image to 3D model） ](https://doc-en.302.ai/247568145e0.md): Image to 3D model from open source program - 根目录 > Image Processing > 302.AI [Pose-Transfer（Human Pose Transformation Result）](https://doc-en.302.ai/252488216e0.md): **Price：0 PTC/call** - 根目录 > Image Processing > 302.AI [Moondream2 (Image Prompt Generation)](https://doc-en.302.ai/330758450e0.md): 302.AI's API comes from models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Image Processing > 302.AI [Pose-Transfer（Human Pose Transformation）](https://doc-en.302.ai/252488247e0.md): Human pose transformation from open source project - 根目录 > Image Processing > 302.AI [Virtual-Tryon （Virtual Clothing V2）](https://doc-en.302.ai/347477580e0.md): **Price: 0.06 PTC/time** - 根目录 > Image Processing > 302.AI [Fetch](https://doc-en.302.ai/207705176e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [SAM（AI-generated mask image）](https://doc-en.302.ai/328205660e0.md): AI-generated MASK Image - 根目录 > Image Processing > 302.AI [image-translate](https://doc-en.302.ai/219451487e0.md): Translate the text in the image into the corresponding language and generate new text - 根目录 > Image Processing > 302.AI [Relight-V2](https://doc-en.302.ai/253670008e0.md): Secondary lighting, IC-Light, which stands for "Imposing Consistent Light," is a project dedicated to manipulating image illumination, project from: https://github.com/lllyasviel/IC-Light - 根目录 > Image Processing > 302.AI [Retouch（Portrait beautification）](https://doc-en.302.ai/328166278e0.md): 302.AI's API comes from models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by ourselves. - 根目录 > Image Processing > 302.AI [Removebg-V3](https://doc-en.302.ai/300886165e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Denoise（AI Denoising）](https://doc-en.302.ai/268505501e0.md): AI Denoising: Remove color noise from photos - 根目录 > Image Processing > 302.AI [Virtual-Tryon（Fetch Result）](https://doc-en.302.ai/252488093e0.md): **Price：0 PTC/call** - 根目录 > Image Processing > 302.AI [Image_Merge](https://doc-en.302.ai/336232655e0.md): Merge Images - 根目录 > Image Processing > 302.AI [HtmltoPng](https://doc-en.302.ai/216464346e0.md): Convert HTML code into PNG format images. - 根目录 > Image Processing > 302.AI [Deblur（AI Deblurring）](https://doc-en.302.ai/268510191e0.md): AI Deblurring - Remove motion blur from photos - 根目录 > Image Processing > 302.AI [Virtual-Tryon](https://doc-en.302.ai/252488177e0.md): Virtual-tryon from open source project - 根目录 > Image Processing > 302.AI [Flux-selfie](https://doc-en.302.ai/222605043e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [image-translate-query](https://doc-en.302.ai/219451488e0.md): Translate the text in the image into the corresponding language and generate new text - 根目录 > Image Processing > 302.AI [Qwen-Image-Edit-Plus (Image Editing Result)](https://doc-en.302.ai/354583212e0.md): **Price: 0 PTC/time** - 根目录 > Image Processing > 302.AI [SvgToPng](https://doc-en.302.ai/216464347e0.md): Convert SVG code into PNG format images. - 根目录 > Image Processing > 302.AI [image-translate-redo](https://doc-en.302.ai/219451489e0.md): Translate the text in the image into the corresponding language and generate new text - 根目录 > Image Processing > 302.AI [Colorize](https://doc-en.302.ai/207705183e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Colorize-V2](https://doc-en.302.ai/207705184e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Removebg](https://doc-en.302.ai/207705186e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Removebg-V2](https://doc-en.302.ai/207705187e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Relight](https://doc-en.302.ai/207705188e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Inpaint](https://doc-en.302.ai/207705189e0.md): Intelligently fill or replace specified areas of an image based on the content of the mask image - 根目录 > Image Processing > 302.AI [Erase](https://doc-en.302.ai/207705190e0.md): Intelligently replace specified areas of an image based on the content of the mask image - 根目录 > Image Processing > 302.AI [Face-swap-V2](https://doc-en.302.ai/207705191e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Relight-background](https://doc-en.302.ai/207705192e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Face-to-many](https://doc-en.302.ai/207705193e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > 302.AI [Llava](https://doc-en.302.ai/207705194e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Image Processing > Vectorizer [Vectorize](https://doc-en.302.ai/207705195e0.md): You can convert ordinary images into infinitely scalable vector graphics using AI. 302.AI API here serves as a demonstration - 根目录 > Image Processing > Stability.ai [Fast Upscale ](https://doc-en.302.ai/222624673e0.md): Our fast zoom service leverages predictive and generative AI to enhance image resolution by up to 4x. This lightweight service, with a processing time of approximately 1 second, is ideal for improving the quality of compressed images, making them suitable for social media posts and other applications. - 根目录 > Image Processing > Stability.ai [Creative Upscale](https://doc-en.302.ai/207705196e0.md): This API can **enlarge images from 64x64 up to 1 million pixels to 4K resolution** with one click. It is characterized by enlarging images by approximately 20-40 times while maintaining the original image quality, and sometimes it can even enhance the image quality. It is best suited for handling images with severe quality loss and is not recommended for photos above 1 million pixels, as it will involve a lot of reimagining (controlled according to the creative ratio). - 根目录 > Image Processing > Stability.ai [Conservative Upscale](https://doc-en.302.ai/207705197e0.md): **Enlarge images from 64x64 pixels to 1 million pixels up to 4K resolution**. More broadly, it can scale images by approximately 20-40 times while preserving details in all aspects. Conservative enlargement minimizes changes to the image and should not be used for reimagining the image. - 根目录 > Image Processing > Stability.ai [Fetch](https://doc-en.302.ai/207705198e0.md): **Fetch task** - 根目录 > Image Processing > Stability.ai [Erase](https://doc-en.302.ai/207705205e0.md): Using image masking technology, unnecessary objects can be removed, such as blemishes in portraits or clutter on a desk. - 根目录 > Image Processing > Stability.ai [Inpaint](https://doc-en.302.ai/207705200e0.md): Intelligently fill or replace the specified area of the image based on the content of the mask image. - 根目录 > Image Processing > Stability.ai [Outpaint](https://doc-en.302.ai/207705201e0.md): Fill in additional content in various directions within the image. Compared to other automatic or manual methods, the Outpaint service minimizes flaws, making the signs of editing on the original image less noticeable. - 根目录 > Image Processing > Stability.ai [Search-and-replace](https://doc-en.302.ai/207705202e0.md): Search-and-replace is a special method of image editing that does not require masking. Users can identify the target to be replaced using simple language through search prompts. The service will automatically recognize and replace objects in the image with the target specified in the search prompt. - 根目录 > Image Processing > Stability.ai [Search-and-recolor](https://doc-en.302.ai/207705203e0.md): Search-and-recolor provides the ability to change the color of specific objects in an image using prompts. This service is a specific version of image retouching that does not require masking. The search and recolor service will automatically segment the objects and recolor them with the color specified in the prompt. - 根目录 > Image Processing > Stability.ai [Remove-background](https://doc-en.302.ai/207705204e0.md): Remove Background service accurately separates the foreground in an image and removes the background. - 根目录 > Image Processing > Stability.ai [Sketch](https://doc-en.302.ai/207705206e0.md): Sketch service offers a perfect solution for design projects that require frequent brainstorming and iteration. It can transform rough hand-drawn sketches into refined outputs, allowing for precise control. For non-sketch images, the service can also utilize the outlines and edges within the image to perform detailed visual adjustments. - 根目录 > Image Processing > Stability.ai [Structure](https://doc-en.302.ai/207705207e0.md): Structure service is capable of generating new images while maintaining the original image structure, making it particularly important in advanced content creation areas such as recreating specific scenes or rendering characters based on models. - 根目录 > Image Processing > Stability.ai [Style](https://doc-en.302.ai/207705208e0.md): Style elements are extracted from the input image (control image) and used to guide the creation of the output image based on prompts. The result is a new image that shares the same style as the control image. - 根目录 > Image Processing > Stability.ai [Replace-Background](https://doc-en.302.ai/238677548e0.md): Replace the image background and readjust the lighting. - 根目录 > Image Processing > Stability.ai [Stable-Fast-3D](https://doc-en.302.ai/216465572e0.md): Convert images to 3D models quickly. - 根目录 > Image Processing > Stability.ai [Stable-Point-3D（Image to 3D Model Conversion -New Version）](https://doc-en.302.ai/253057122e0.md): Converting an image to a 3D model using the point cloud method. Introduction：https://stability.ai/news/stable-point-aware-3d - 根目录 > Image Processing > Stability.ai [Style-Transfer ](https://doc-en.302.ai/326232198e0.md): Style transfer applies the visual features of a reference style image to a target image. Style guidance, on the other hand, extracts style elements from an input image (control image) and uses these elements to guide the generation of output images based on prompts; style transfer specifically transforms existing content while preserving the original composition. This tool can help create consistent content across multiple assets. - 根目录 > Image Processing > Glif [Glif（Portrait Photo Stylization）](https://doc-en.302.ai/207705209e0.md): Upload a portrait photo and select a style filter for generation. - 根目录 > Image Processing > Glif [Glif（Photo-to-Sculpture）](https://doc-en.302.ai/207705211e0.md): Upload a photo and convert it into a sculpture. - 根目录 > Image Processing > Glif [Glif（Photo Pixelation）](https://doc-en.302.ai/207705212e0.md): Upload a photo and convert it into a Pixelation. - 根目录 > Image Processing > Glif [Glif（Logo Materialization）](https://doc-en.302.ai/207705210e0.md): Upload a logo image, select the desired material, and transform the logo. Example material: diamond - 根目录 > Image Processing > Glif [Glif（Image-to-GIF）](https://doc-en.302.ai/207705213e0.md): Upload a photo, animate the image using AI, and generate a GIF. - 根目录 > Image Processing > Clipdrop [Cleanup](https://doc-en.302.ai/207705214e0.md): Clipdrop is a company that provides AI image editing services, allowing for quick and easy modifications to images. We are fully aligned with their official API, and you only need to **replace the API Base URL to use it**. - 根目录 > Image Processing > Clipdrop [Upscale](https://doc-en.302.ai/207705215e0.md): Clipdrop is a company that provides AI image editing services, allowing for quick and easy modifications to images. We are fully aligned with their official API, and you only need to replace the API Base URL to use it. - 根目录 > Image Processing > Clipdrop [Remove-background](https://doc-en.302.ai/207705216e0.md): Clipdrop is a company that provides AI image editing services, allowing for quick and easy modifications to images. We are fully aligned with their official API, and you only need to replace the API Base URL to use it. - 根目录 > Image Processing > Clipdrop [Uncrop](https://doc-en.302.ai/207705217e0.md): Clipdrop is a company that provides AI image editing services, allowing for quick and easy modifications to images. We are fully aligned with their official API, and you only need to replace the API Base URL to use it. - 根目录 > Image Processing > Recraft [Vectorize Image](https://doc-en.302.ai/231458786e0.md): The latest mysterious model raising eyebrows on the LLM Big Model list is Red_Panda.，from: https://www.recraft.ai/ - 根目录 > Image Processing > Recraft [Remove Background](https://doc-en.302.ai/231458787e0.md): The latest mysterious model raising eyebrows on the LLM Big Model list is Red_Panda.，from: https://www.recraft.ai/ - 根目录 > Image Processing > Recraft [Clarity Upscale](https://doc-en.302.ai/231458788e0.md): The latest mysterious model raising eyebrows on the LLM Big Model list is Red_Panda.，from: https://www.recraft.ai/ - 根目录 > Image Processing > Recraft [Generative Upscale](https://doc-en.302.ai/231458789e0.md): The latest mysterious model raising eyebrows on the LLM Big Model list is Red_Panda.，from: https://www.recraft.ai/ - 根目录 > Image Processing > BRIA [Remove Background](https://doc-en.302.ai/235057906e0.md): Remove the background of an image，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Blur Background](https://doc-en.302.ai/235057907e0.md): Blur the image background，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Generate Background](https://doc-en.302.ai/235057904e0.md): Select the image subject and regenerate the background，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Erase Foreground](https://doc-en.302.ai/235057894e0.md): Erase the image foreground, leaving only the background，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Eraser](https://doc-en.302.ai/235057905e0.md): Erase the selected part of the image，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Expand Image](https://doc-en.302.ai/235057903e0.md): Extend the image boundaries using AI to imagine the rest，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Increase Resolution](https://doc-en.302.ai/235057902e0.md): Increase the image resolution，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Crop](https://doc-en.302.ai/235057895e0.md): Automatically crop the subject part of image，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Cutout](https://doc-en.302.ai/235057897e0.md): Remove the background of the product image and crop it out，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Packshot](https://doc-en.302.ai/235057898e0.md): Convert the product image into a close-up，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Shadow](https://doc-en.302.ai/235057899e0.md): Generate a shadow for the product image，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Scene](https://doc-en.302.ai/235057900e0.md): Generate a scene for the product image，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Caption](https://doc-en.302.ai/235057901e0.md): Obtain image description，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Register](https://doc-en.302.ai/235057896e0.md): Upload the image for further editing，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Mask](https://doc-en.302.ai/236117381e0.md): Split the image into different sections and generate a compressed package,from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Presenter info](https://doc-en.302.ai/236117275e0.md): Analyze facial information，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Modify Presenter](https://doc-en.302.ai/236117179e0.md): Edit facial details，from: https://bria.ai/ - 根目录 > Image Processing > BRIA [Delayer Image](https://doc-en.302.ai/236117090e0.md): Convert image to multi-layer PSD，from: https://bria.ai/ - 根目录 > Image Processing > Flux > Official API [Generate（Image Edit）](https://doc-en.302.ai/304611655e0.md): Official Documentation：https://api.bfl.ml/scalar#tag/tasks/POST/v1/flux-kontext-pro - 根目录 > Image Processing > Flux > Official API [Result（Query Task）](https://doc-en.302.ai/304611656e0.md): **Price: Free** - 根目录 > Image Processing > Flux [Flux-V1.1-Ultra-Redux（Image-to-image generation-Ultra）](https://doc-en.302.ai/236383171e0.md): Given an input image, FLUX.1 Redux can reproduce images with slight variations, allowing for the refinement of the given image. - 根目录 > Image Processing > Flux [Flux-V1.1-Pro-Redux（Image-to-image generation-Pro）](https://doc-en.302.ai/236383172e0.md): Given an input image, FLUX.1 Redux can reproduce images with slight variations, allowing for the refinement of the given image. - 根目录 > Image Processing > Flux [Flux-Dev-Redux（Image-to-image generation-Dev）](https://doc-en.302.ai/236383176e0.md): Given an input image, FLUX.1 Redux can reproduce images with slight variations, allowing for the refinement of the given image. - 根目录 > Image Processing > Flux [Flux-Schnell-Redux（Image-to-image generation-Schnell）](https://doc-en.302.ai/236383177e0.md): Given an input image, FLUX.1 Redux can reproduce images with slight variations, allowing for the refinement of the given image. - 根目录 > Image Processing > Flux [Flux-V1-Pro-Canny（Object consistency）](https://doc-en.302.ai/236383173e0.md): Structural adjustment uses intelligent edges to maintain precise control during image transformation. By preserving the structure of the original image through edge maps, users can perform text-guided edits while keeping the core composition intact. This is particularly effective for retexturing images. - 根目录 > Image Processing > Flux [Flux-V1-Pro-Depth（Depth consistency）](https://doc-en.302.ai/236383174e0.md): Structural adjustment uses depth detection to maintain precise control during image transformation. By preserving the structure of the original image through depth maps, users can perform text-guided edits while keeping the core composition intact. This is particularly effective for retexturing images. - 根目录 > Image Processing > Flux [Flux-V1-Pro-Fill（Partial repainting）](https://doc-en.302.ai/236383175e0.md): Structural adjustment uses depth detection to maintain precise control during image transformation. By preserving the structure of the original image through depth maps, users can perform text-guided edits while maintaining the integrity of the core composition. This is particularly effective for retexturing images. - 根目录 > Image Processing > Flux [Flux-Kontext-Pro（Image Edit）](https://doc-en.302.ai/303445344e0.md): Given an input image, edit the image, the latest FLUX.1 Kontext model - 根目录 > Image Processing > Flux [Flux-Kontext-Max（Image Edit）](https://doc-en.302.ai/303446055e0.md): Given an input image, edit the image, the latest FLUX.1 Kontext model - 根目录 > Image Processing > Flux [Flux-Kontext-Dev（Image Edit）](https://doc-en.302.ai/325627969e0.md): Given an input image, edit the picture using the latest FLUX.1 Kontext model. - 根目录 > Image Processing > Hyper3D [Hyper3d-Rodin（Generate 3D models）](https://doc-en.302.ai/240592861e0.md): Image-to-3D model generation from Hyper3D, capable of creating ultra-detailed 3D models. - 根目录 > Image Processing > Hyper3D [Hyper3d-Rodin（Obtain task results）](https://doc-en.302.ai/242657673e0.md): Video generation model from Tencent Open Source - 根目录 > Image Processing > Tripo3D [Task（Task Submission）](https://doc-en.302.ai/243969346e0.md): For specific usage, please refer to the official documentation: - 根目录 > Image Processing > Tripo3D [Upload（Image Upload）](https://doc-en.302.ai/243969345e0.md): **Price：0 PTC/call** - 根目录 > Image Processing > Tripo3D [Fetch](https://doc-en.302.ai/243969347e0.md): **Price：0 PTC/call** - 根目录 > Image Processing > FASHN [Fashn-Tryon（Virtual Try-On）](https://doc-en.302.ai/244116847e0.md): Virtual Try-On from FASHN - 根目录 > Image Processing > FASHN [Fashn-Tryon（Virtual Try-On v1.5）](https://doc-en.302.ai/292523223e0.md): Virtual Try-On from FASHN - 根目录 > Image Processing > Ideogram [Edit（subject reference）](https://doc-en.302.ai/337855625e0.md): Text-to-image generation from Ideogram excels at producing accurate text and posters. It now supports inputting portrait subjects to generate images referencing the subject. - 根目录 > Image Processing > Ideogram [Remix（subject reference）](https://doc-en.302.ai/337855626e0.md): Text-to-image generation from Ideogram excels at producing accurate text and posters. It now supports inputting portrait subjects to generate images referencing the subject. - 根目录 > Image Processing > Ideogram [Edit（Image EditionV3）](https://doc-en.302.ai/292512023e0.md): Text-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Ideogram [Remix（Image to ImageV3）](https://doc-en.302.ai/292848604e0.md): Text-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Ideogram [Reframe（Image ExtensionV3）](https://doc-en.302.ai/292849718e0.md): Text-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Ideogram [Replace Background（V3）](https://doc-en.302.ai/292850289e0.md): Text-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Ideogram [Remix（Image to Image）](https://doc-en.302.ai/245824806e0.md): Image-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Ideogram [Upscale（Image Upscaling）](https://doc-en.302.ai/245824807e0.md): Double the size of the Ideogram image - 根目录 > Image Processing > Ideogram [Describe（Image Description）](https://doc-en.302.ai/245824808e0.md): Image description from Ideogram - 根目录 > Image Processing > Ideogram [Edit（Image Edition）](https://doc-en.302.ai/245824809e0.md): Image-to-image generation from Ideogram, with its key advantage being the ability to generate accurate text and posters. - 根目录 > Image Processing > Doubao [SeedEdit（Image Command Editing） ](https://doc-en.302.ai/246593603e0.md): Image Editing Model from Doubao - 根目录 > Image Processing > Doubao [Character（Character Feature Preservation） ](https://doc-en.302.ai/246593604e0.md): Image generation model from SeedEdit - 根目录 > Image Processing > Doubao [SeedEdit_v3.0 (Image Command Editing)](https://doc-en.302.ai/318268273e0.md): Image Generation Model from Doubao - 根目录 > Image Processing > Doubao [SeedEdit_v3.0 (Result Acquisition)](https://doc-en.302.ai/318268274e0.md): seededit_v3.0 Task Acquisition - 根目录 > Image Processing > Doubao [Portrait (Portrait Photography)](https://doc-en.302.ai/318268275e0.md): Image-to-Image 3.0 - Portrait Photography - 根目录 > Image Processing > Doubao [Portrait (Result Acquisition)](https://doc-en.302.ai/318268276e0.md): Image-to-Image 3.0 - Portrait Photography Task Acquisition - 根目录 > Image Processing > Doubao [Seed3D（Image generation 3D model）](https://doc-en.302.ai/371028194e0.md): Create 3D models from JiMeng. The model will generate 3D based on the provided image information. Once the generation is complete, you can query tasks by conditions and obtain the generated 3D file. - 根目录 > Image Processing > Doubao [Seed3D (Get Task Results)](https://doc-en.302.ai/371028195e0.md): **Price: 0 PTC/time** - 根目录 > Image Processing > Kling [Virtual-Try-On](https://doc-en.302.ai/251317721e0.md): Virtual Try-On by Kling - 根目录 > Image Processing > Kling [Fetch（Get Task Result） ](https://doc-en.302.ai/251317722e0.md): Virtual Try-On by Kling - 根目录 > Image Processing > Kling [Images-expand](https://doc-en.302.ai/322635848e0.md): From Kling Expansion - 根目录 > Image Processing > Kling [Fetch（Retrieve the results of the image upscaling task） ](https://doc-en.302.ai/322722499e0.md): From Kling Raw Images - 根目录 > Image Processing > StepFun [Step1x-Edit（Modify Image）](https://doc-en.302.ai/293006686e0.md): Image modification from StepFun, simply input the image and prompt words to modify the image - 根目录 > Image Processing > Bagel [Bagel-Edit（Image Edit）](https://doc-en.302.ai/300866520e0.md): Image model from Byte Open Source, similar to gpt-image-1, can generate images and modify them - 根目录 > Image Processing > Gongji Computing > Flux Dev [Create flux_dev text-to-image task ](https://doc-en.302.ai/326232199e0.md): Create flux_dev text-to-image task - 根目录 > Image Processing > Gongji Computing > Flux Dev [Query flux_dev text-to-image task](https://doc-en.302.ai/326232200e0.md): Create flux_dev text-to-image task - 根目录 > Image Processing > Gongji Computing > Flux Kontext Dev [Create flux_kontext_dev image editing task](https://doc-en.302.ai/326232201e0.md): Create flux_kontext_dev image editing task - 根目录 > Image Processing > Gongji Computing > Flux Kontext Dev [Query flux_kontext_dev image editing tasks](https://doc-en.302.ai/326232202e0.md): View flux_kontext_dev image editing task - 根目录 > Image Processing > Gongji Computing > Flux Kontext Dev [Create LoRA Image Editing Task](https://doc-en.302.ai/330758451e0.md): Create flux_kontext_dev+lora image editing task - 根目录 > Image Processing > Gongji Computing > Flux Kontext Dev [View LoRA Image Editing Task](https://doc-en.302.ai/330846705e0.md): View flux_kontext_dev+lora image editing tasks - 根目录 > Image Processing > Gongji Computing > Face Swapper [Create face_swapper task](https://doc-en.302.ai/326232203e0.md): Create face_swapper task (replace the original face with the target face) - 根目录 > Image Processing > Gongji Computing > Face Swapper [Get face_swapper task](https://doc-en.302.ai/326232204e0.md): Query face_swapper task - 根目录 > Image Processing > Gongji Computing > Clothes Changer [Create clothes changer task without mask](https://doc-en.302.ai/326232205e0.md): Create Unmasked Outfit Change Task - 根目录 > Image Processing > Gongji Computing > Clothes Changer [Get clothes changer task without mask](https://doc-en.302.ai/326232206e0.md): View unmasked outfit change task - 根目录 > Image Processing > Gongji Computing > Clothes Changer [Create clothes changer task with mask](https://doc-en.302.ai/326232207e0.md): Create Masked Dress-up Task - 根目录 > Image Processing > Gongji Computing > Clothes Changer [Get clothes changer task with mask](https://doc-en.302.ai/326232208e0.md): View masked outfit change task - 根目录 > Image Processing > Gongji Computing > Anything Changer [创建无遮罩换任意物品任务](https://doc-en.302.ai/337270288e0.md): Create Unmasked Arbitrary Item Replacement Task - 根目录 > Image Processing > Gongji Computing > Anything Changer [查看无遮罩换任意物品任务](https://doc-en.302.ai/337270289e0.md): View unmasked swap any item task - 根目录 > Image Processing > Gongji Computing > Anything Changer [创建有遮罩换任意物品任务](https://doc-en.302.ai/337270290e0.md): Create Masked Arbitrary Item Replacement Task - 根目录 > Image Processing > Gongji Computing > Anything Changer [查看有遮罩换任意物品任务](https://doc-en.302.ai/337270291e0.md): View masked swap any item task - 根目录 > Image Processing > Gongji Computing > style_transfer [创建风格迁移任务](https://doc-en.302.ai/337270294e0.md): Create Style Transfer Task - 根目录 > Image Processing > Gongji Computing > style_transfer [查看风格迁移任务](https://doc-en.302.ai/337270295e0.md): View Style Transfer Task - 根目录 > Image Processing > Gongji Computing > Image Eliminater [Create image removal task](https://doc-en.302.ai/342849000e0.md): Create Image Removal Task - 根目录 > Image Processing > Gongji Computing > Image Eliminater [View image removal task](https://doc-en.302.ai/342849001e0.md): View Image Elimination Task - 根目录 > Image Processing > Image2Reality [Create Anime-to-Real Task](https://doc-en.302.ai/337270292e0.md): Create Anime-to-Realistic Task - 根目录 > Image Processing > Image2Reality [View Anime-to-Real Tasks](https://doc-en.302.ai/337270293e0.md): View Anime Task as Live-action Task - 根目录 > Image Processing > Hunyuan3D [Hunyuan3d-v21 (Generate 3D Model)](https://doc-en.302.ai/329405862e0.md): From Huanyuan: Image-to-3D Model Generation Version 2.1, capable of producing ultra-detailed 3D models - 根目录 > Image Processing > Hunyuan3D [Hunyuan3d-v21 (Get Task Result)](https://doc-en.302.ai/329405863e0.md): **Price: 0 PTC/time** - 根目录 > Image Processing > Hidream [Hidream-E1 (Image Editing)](https://doc-en.302.ai/329405864e0.md): Image Editing Model from Zhixiang Future - 根目录 > Image Processing > Tongyi Wanxiang [Qwen-Image-Edit-Plus（Alibaba Cloud Deployment） ](https://doc-en.302.ai/400586401e0.md): Tongyi Qianwen - Image Editing Model (qwen-image-edit-plus-2025-12-15) supports multi-image input and multi-image output, enabling precise modification of text within images, addition, deletion, or movement of objects, alteration of subject actions, migration of image styles, and enhancement of image details. - 根目录 > Image Processing > Tongyi Wanxiang [Qwen-Image-Edit (Alibaba Cloud Deployment)](https://doc-en.302.ai/340677698e0.md): Qwen-Image-Edit supports precise bilingual text editing (Chinese and English), color adjustment, detail enhancement, style transfer, object addition and removal, changing positions and actions, and other operations, enabling complex image and text editing. - 根目录 > Image Processing > Tongyi Wanxiang [Wanx2.1-ImageEdit (Image Editing)](https://doc-en.302.ai/340677699e0.md): This model enables diverse image editing through simple instructions, suitable for scenarios such as image extension, watermark removal, style transfer, image restoration, and image enhancement. Currently, the following features are supported: - 根目录 > Image Processing > Tongyi Wanxiang [Wan2.5-i2i-Preview (Image Editing)](https://doc-en.302.ai/354583213e0.md): This model enables diverse image editing through simple instructions, suitable for scenarios such as image extension, watermark removal, style transfer, image restoration, and image enhancement. Currently, the following features are supported: - 根目录 > Image Processing > Tongyi Wanxiang [Qwen-MT-Image (Image Translation)](https://doc-en.302.ai/342267511e0.md): The Tongyi Qianwen Image Translation Model (Qwen-MT-Image) supports translating text from images in 11 languages into Chinese or English, accurately preserving original layout and content information, and providing customizable features such as terminology definition, sensitive word filtering, and image subject detection. - 根目录 > Image Processing > Tongyi Wanxiang [Tasks (Get Task Results)](https://doc-en.302.ai/340677700e0.md): Get Task Result - 根目录 > Image Processing > Topazlabs [Sharpen](https://doc-en.302.ai/356350969e0.md): With our advanced sharpening model, you can restore clarity and vivid detail to images that appear blurry or slightly out of focus. By analyzing edges, textures, and details, our AI-powered sharpening endpoint intelligently enhances sharpness without introducing unwanted artifacts. Whether your image is affected by camera shake, slight motion blur, or lens softness, Topaz Labs' sharpening model helps you easily recover more fine details. - 根目录 > Image Processing > Topazlabs [Sharpen Generation](https://doc-en.302.ai/360266556e0.md): With our advanced sharpening model, images that appear blurry or slightly out of focus can be restored to clarity and vivid detail. By analyzing edges, textures, and details, our AI-powered sharpening endpoint intelligently enhances sharpness without introducing unwanted artifacts. Whether your image is affected by camera shake, slight motion blur, or lens softness, Topaz Labs' sharpening model helps you effortlessly recover more fine details. - 根目录 > Image Processing > Topazlabs [Enhance](https://doc-en.302.ai/356350970e0.md): Enhance images by denoising, sharpening, and upscaling while preserving image details, including important objects such as faces. - 根目录 > Image Processing > Topazlabs [Enhance Generation](https://doc-en.302.ai/360266557e0.md): Enhance images by denoising, sharpening, and upscaling while preserving image details, including faces of important objects. - 根目录 > Image Processing > Topazlabs [Denoise](https://doc-en.302.ai/356350971e0.md): Our denoising model efficiently removes grain and high ISO noise from photos, ensuring smooth images with clear details. This endpoint is especially useful for processing low-light photography or old photos scanned from film, as it significantly reduces unwanted noise or artifacts while preserving important image structures. - 根目录 > Image Processing > Topazlabs [Restore](https://doc-en.302.ai/356350972e0.md): Below is our restoration model for enhancing older and low-quality photos. - 根目录 > Image Processing > Topazlabs [Lighting](https://doc-en.302.ai/356350973e0.md): Below is our lighting model, which can be used to adjust and balance your images, enhancing their quality even under poor original lighting conditions. - 根目录 > Image Processing > Topazlabs [Get Task Results](https://doc-en.302.ai/356350974e0.md): **Price: Free** - 根目录 > Image Processing > Topazlabs [Download](https://doc-en.302.ai/356350975e0.md): Download input or processed images via a pre-signed URL. This API allows you to securely access input or processed images through a time-limited URL. - 根目录 > Image Processing > Wavespeed > General Interface [Wavespeed Generation Task](https://doc-en.302.ai/360313609e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 at the same price. - 根目录 > Image Processing > Wavespeed > General Interface [Wavespeed Retrieval Task](https://doc-en.302.ai/360313610e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Image Processing > Wavespeed [runwayml/gen4-image-turbo](https://doc-en.302.ai/365103924e0.md): Gen-4 Image Turbo is cheaper than Gen-4 Image and offers a 2.5x speed increase. It is an image model with reference image support, allowing you to use up to 3 reference images to create the exact image you want, capturing every angle. - 根目录 > Image Processing > Wavespeed [wavespeed-ai/image-captioner](https://doc-en.302.ai/383603113e0.md): image-captioner is a high-precision image caption generator that can produce detailed, human-like descriptions from images. It is ideal for content understanding, accessibility, dataset annotation, SEO, and multimodal AI workflows. Ready-to-use REST inference API with optimal performance, no cold start, and affordable pricing. - 根目录 > Image Processing > Photoroom [Remove Background](https://doc-en.302.ai/405054067e0.md): **🛠️ API Function** - 根目录 > Video Generation > Unified Interface > 302 Format V1 [Create Video Generation Task](https://doc-en.302.ai/275315909e0.md): This API is designed to integrate the generation function interfaces provided by various deployers, extract commonly used parameter fields and response fields, standardize field naming and parameter/response data formats, and improve usage efficiency. - 根目录 > Video Generation > Unified Interface > 302 Format V1 [Get Video Task Info](https://doc-en.302.ai/275315946e0.md): This API is mainly responsible for retrieving the generated video, storing it in the 302 file system, and returning a public network link as the response. For additional information, you can refer to the data field. - 根目录 > Video Generation > Unified Interface > 302 Format V2 [Create Video Generation Task](https://doc-en.302.ai/343760431e0.md): Model list and model-supported feature documentation: - 根目录 > Video Generation > Unified Interface > 302 Format V2 [Get Video Task Info](https://doc-en.302.ai/343760432e0.md): - 根目录 > Video Generation > Unified Interface > 302 Format V2 [ Webhook Request Example on Success](https://doc-en.302.ai/343760433e0.md): Any response with a 200 status code is considered a successful send. - 根目录 > Video Generation > 302.AI [Image-to-video ](https://doc-en.302.ai/207705233e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Video Generation > 302.AI [Live-portrait](https://doc-en.302.ai/207705234e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Video Generation > 302.AI [Video-To-Video](https://doc-en.302.ai/230622642e0.md): Transforming a video into another style with cue words - 根目录 > Video Generation > 302.AI [Fetch](https://doc-en.302.ai/207705232e0.md): 302.AI's API is derived from models we've deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Video Generation > 302.AI [Latentsync (Open source digital person)](https://doc-en.302.ai/258261826e0.md): ByteDance's open source digital person realizes lip-synchronization of video and voice. - 根目录 > Video Generation > 302.AI [Latentsync (get task results)](https://doc-en.302.ai/258261827e0.md): ByteDance's open source digital person realizes lip-synchronization of video and voice - 根目录 > Video Generation > 302.AI [Upscale-Video（Video Enhancement）](https://doc-en.302.ai/290751164e0.md): 302.AI's API is derived from the models we deploy ourselves on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Video Generation > 302.AI [Upscale-Video（Get Video Results）](https://doc-en.302.ai/290751435e0.md): The API of 302.AI is derived from the models we have deployed ourselves on cloud GPUs. Some of the models are open-source, while others are fine-tuned or developed by us. - 根目录 > Video Generation > 302.AI [Stable-Avatar (Open Source Digital Human)](https://doc-en.302.ai/340677707e0.md): 302.AI's API comes from models we have deployed on our own cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Video Generation > 302.AI [Stable-Avatar (Get Task Results)](https://doc-en.302.ai/340677708e0.md): 302.AI's API is powered by models we have deployed on our own cloud GPUs. Some models are open-source, while others have been fine-tuned or developed by us. - 根目录 > Video Generation > 302.AI [Wan-2.2-i2v-fast (Wan2.2 Fast Version)](https://doc-en.302.ai/354942233e0.md): 302AI self-hosted **Wan2.2 Fast Version** supports image-to-video generation (first frame required) and end-frame constraints. It comes with built-in LoRA extension, random seed control, and automatic frame interpolation up to 30FPS. You can flexibly set the number of frames, resolution (480p/720p), and frame rate, meeting the rapid generation needs for creative videos, motion demos, and short advertisements. - 根目录 > Video Generation > 302.AI [Wan-2.2-i2v-fast (Get Video Result)](https://doc-en.302.ai/354942234e0.md): The API of 302.AI comes from models we have deployed on cloud GPU. Some models are open-source, while others are fine-tuned or developed by ourselves. - 根目录 > Video Generation > 302.AI-ComfyUI [Create Video Face Swap Task](https://doc-en.302.ai/289966991e0.md): **Price: 0.03 PTC/second** - 根目录 > Video Generation > 302.AI-ComfyUI [Query Video Face Swap Task Status](https://doc-en.302.ai/289967053e0.md): **Price：free** - 根目录 > Video Generation > 302.AI-ComfyUI [Create Video Face Swap Task (V2)](https://doc-en.302.ai/337721027e0.md): **Price: 0.03 PTC/second based on the length of the input video** - 根目录 > Video Generation > 302.AI-ComfyUI [Query Video Face Swap Task Status (V2)](https://doc-en.302.ai/337721028e0.md): **Price: Free** - 根目录 > Video Generation > Stable Diffusion [Image-to-video](https://doc-en.302.ai/207705235e0.md): Generate a short video based on an initial image using Stable Video Diffusion. - 根目录 > Video Generation > Stable Diffusion [Fetch Image-to-video](https://doc-en.302.ai/207705236e0.md): Generate a short video based on an initial image using Stable Video Diffusion. - 根目录 > Video Generation > Luma AI [Submit（Text / Image to Video）](https://doc-en.302.ai/207705237e0.md): Text/Image-to-Video generation from Luma AI. Generate a 5-second video based on text or images. - 根目录 > Video Generation > Luma AI [Extend（Video）](https://doc-en.302.ai/207705238e0.md): Extend（Video） from Luma AI. This API is used to extend previously generated videos, adding 5s of playback time. - 根目录 > Video Generation > Luma AI [Fetch](https://doc-en.302.ai/207705239e0.md): Luma AI：https://lumalabs.ai/dream-machine - 根目录 > Video Generation > Runway [Submit（Text to Video）](https://doc-en.302.ai/207705240e0.md): **Text-to-video from Runway Gen-3, creating a 10-second video based on the text**. - 根目录 > Video Generation > Runway [Submit（Image to Video）](https://doc-en.302.ai/207705241e0.md): **Text-to-video from Runway Gen-3, creating a 10-second video based on the image**. - 根目录 > Video Generation > Runway [Submit（Image to Video Rapid）](https://doc-en.302.ai/207705242e0.md): **Runway Gen-3 Turbo**, updated on August 15th, allows for **rapid video generation from images**, but currently does not support text-to-video generation. - 根目录 > Video Generation > Runway [Submit（Image-to-Video Generation with Gen4）](https://doc-en.302.ai/279710322e0.md): The updated Runway Gen4 (as of April 1) allows fast video generation from images. Text-to-video is not supported at the moment. - 根目录 > Video Generation > Runway [Submit（Image to Video Generation Gen4-Turbo）](https://doc-en.302.ai/282193420e0.md): Runway Gen4 Turbo updated on April 1st, allows quick image-to-video generation, currently does not support text-to-video generation. - 根目录 > Video Generation > Runway [Submit（Video to Video）](https://doc-en.302.ai/223500418e0.md): The raw video from Runway Gen-3 transforms the original footage into various styles based on the video content and cue words. - 根目录 > Video Generation > Runway [Submit（Video to Video Rapid）](https://doc-en.302.ai/224541525e0.md): **Runway Gen-3 Turbo**, updated on August 15th, allows for **rapid video generation from images**, but currently does not support text-to-video generation. - 根目录 > Video Generation > Runway [Act-two（Video Style Transfer）](https://doc-en.302.ai/230623447e0.md): ### Runway’s Latest Generative Motion Capture Tool — Act-Two: Instantly Turn Videos into Animation - 根目录 > Video Generation > Runway [Submit（Video extension） ](https://doc-en.302.ai/242641700e0.md): Expand a landscape video to portrait or expand a portrait video to landscape - 根目录 > Video Generation > Runway [Aleph (Video Editing)](https://doc-en.302.ai/330758449e0.md): Runway Aleph is an advanced context-aware video editing model that sets a new frontier for multi-task visual generation. It is capable of extensive edits to input videos, such as adding, removing, and transforming objects, generating any angle of a scene, as well as modifying style and lighting, among others. - 根目录 > Video Generation > Runway [Fetch](https://doc-en.302.ai/207705243e0.md): **Runway Gen-3, creating a 10-second video based on the text/image**. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.0 Rapid-5s）](https://doc-en.302.ai/207705247e0.md): **Image-to-video** generated by Kling, **a quick 5-second version**. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.0 Rapid-10s）](https://doc-en.302.ai/222607629e0.md): **Image-to-video** generated by Kling, **a quick 10-second version**. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.5 Rapid-5s） ](https://doc-en.302.ai/238109409e0.md): **Image-to-video** generated by Kling, **a quick 5-second version**. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.5 Rapid-10s） ](https://doc-en.302.ai/238109410e0.md): **Image-to-video** generated by Kling, **a quick 10-second version**. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video_HQ（Image to Video 1.5 HQ-5s）](https://doc-en.302.ai/207705246e0.md): **The Tucson video from Kling, in the 5-second HD format, is automatically upgraded to Kling-1.5.** - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video_HQ（Image to Video 1.5 HQ-10s）](https://doc-en.302.ai/222607628e0.md): The Tucson video from Kling, in the 10-second HD format, is automatically upgraded to Kling-1.5. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.6 Standard-5s）](https://doc-en.302.ai/246824589e0.md): Image-to-video generated by Kling, a standard 5-second version. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.6 Standard-10s）](https://doc-en.302.ai/246824622e0.md): Image-to-video generated by Kling, a standard 10-second version. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Multiple Images Reference）](https://doc-en.302.ai/321066087e0.md): With the multi-image reference feature from Kling, you can upload up to 4 images. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.6 HQ-5s）](https://doc-en.302.ai/246824653e0.md): Image-to-video generated by Kling, a HQ 5-second version. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 1.6 HQ-10s）](https://doc-en.302.ai/246824698e0.md): Image-to-video generated by Kling, a HQ 10-second version. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video 2.0 – HD – 5s）](https://doc-en.302.ai/284770464e0.md): - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video 2.0 – HD – 10s）](https://doc-en.302.ai/284770552e0.md): This is the 5-second HD version of Kling’s Image-to-Video 2.0. - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image video 2.1-5 seconds）](https://doc-en.302.ai/303458496e0.md): From Kling's Tusheng Video 2.1 version, 5 seconds version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image video 2.1-10 seconds）](https://doc-en.302.ai/303458497e0.md): From Kling's Tusheng Video 2.1, 10-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image Video 2.1-HD-10 seconds）](https://doc-en.302.ai/303458499e0.md): From Kling's Tusheng video version 2.1, high-definition 10-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image Video 2.1-HD-5 seconds）](https://doc-en.302.ai/303458498e0.md): From Kling's Tusheng video version 2.1, high-definition 5-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 2.1–Master –5seconds） ](https://doc-en.302.ai/321061920e0.md): From Kling Video Generation 2.1 Master Edition, HD 5-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image to Video 2.1–Master–10seconds） ](https://doc-en.302.ai/321061921e0.md): From Kling Video Generation 2.1 Master Edition, HD 10-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video2.5-Turbo-5S)](https://doc-en.302.ai/354444882e0.md): From Kling's Turbo 2.5 version for image-to-video, HD 5-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video2.5-Turbo-10s）](https://doc-en.302.ai/354444883e0.md): From Kling's Turbo 2.5 Image-to-Video, HD 10-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video2.5-Turbo-Std-5s)](https://doc-en.302.ai/376682635e0.md): From Kling's Turbo 2.5 Image-to-Video, standard 5-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Image-to-Video2.5-Turbo-Std-10s)](https://doc-en.302.ai/376682636e0.md): From Kling's Turbo 2.5 Image-to-Video, standard 10-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video（Kling-o1）](https://doc-en.302.ai/386885928e0.md): From Kling (Kling o1) - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video (Image to Video 2.6-5 seconds)](https://doc-en.302.ai/387052030e0.md): From Kling's Image to Video Version 2.6, 5-second version - 根目录 > Video Generation > Kling > 302 format > Image to Video [Image2Video (Image to Video 2.6-10 seconds)](https://doc-en.302.ai/387053053e0.md): From Kling's Image to Video Version 2.6, 5-second version - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text to Video 1.0 Rapid-5s）](https://doc-en.302.ai/207705249e0.md): Text-to-video generated by Kling, a quick 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video_HQ（Text to Video 1.5 HQ-5s）](https://doc-en.302.ai/207705248e0.md): Text-to-video generated by Kling, **a High-quality 5-second version**. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video_HQ（Text to Video 1.5 HQ-10s）](https://doc-en.302.ai/207705245e0.md): Text-to-video generated by Kling, **a High-quality 10-second version**. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text to Video 1.6 Standard-5s） ](https://doc-en.302.ai/246827254e0.md): Text-to-video generated by Kling, a standard 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text to Video 1.6 Standard-10s） ](https://doc-en.302.ai/246824267e0.md): Text-to-video generated by Kling, a standard 10-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text to Video 1.6 HQ-5s） ](https://doc-en.302.ai/246824453e0.md): Text-to-video generated by Kling, a High-quality 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text to Video 1.6 HQ-10s） ](https://doc-en.302.ai/246824516e0.md): Text-to-video generated by Kling, a High-quality 10-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-to-Video 2.0 – HD – 5s） ](https://doc-en.302.ai/284770440e0.md): This is the 5-second HD version of Kling’s Text-to-Video 2.0. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-to-Video 2.1 – Master Edition – 5S）](https://doc-en.302.ai/321054271e0.md): From Kling Video Generation 2.1 Master Edition, HD 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-to-Video 2.1 – Master Edition – 10S）](https://doc-en.302.ai/321054272e0.md): From Kling Video Generation 2.1 Master Edition. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-To-Video-2.5-Turbo-5S）](https://doc-en.302.ai/354446476e0.md): From Kling's WenSheng Video 2.5 Turbo version, HD 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-To-Video-2.5-Turbo-10S） ](https://doc-en.302.ai/354446477e0.md): From Kling's Text-to-Video 2.5 Turbo version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-to-Video-2.5-Turbo-Std-5s）](https://doc-en.302.ai/376725818e0.md): From Kling's Turbo 2.5 Text-to-Video, standard 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video（Text-to-Video-2.5-Turbo-Std-10s）](https://doc-en.302.ai/376725819e0.md): From Kling's Turbo 2.5 Text-to-Video, standard 10-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video (Text-to-Video 2.6--5 seconds)](https://doc-en.302.ai/387066794e0.md): From Kling's Text-to-Video version 2.6, 5-second version. - 根目录 > Video Generation > Kling > 302 format > Text to Video [Txt2Video (Text-to-Video 2.6 -- 10 seconds)](https://doc-en.302.ai/387067150e0.md): From Kling's Text-to-Video Version 2.6, 10-second version. - 根目录 > Video Generation > Kling > 302 format [Extend_Video](https://doc-en.302.ai/223198280e0.md): The extended video from Kling allows for 5-second extensions, but the HD version does not support this feature. - 根目录 > Video Generation > Kling > 302 format [Fetch](https://doc-en.302.ai/207705244e0.md): **Image-to-video and Text-to-video generated by Kling.** - 根目录 > Video Generation > Kling > Official format [Text2video (Text-to-Video Official API)](https://doc-en.302.ai/305983163e0.md): From Kling official API - 根目录 > Video Generation > Kling > Official format [Text2video (Text-to-Video Get Task Result)](https://doc-en.302.ai/305983164e0.md): From Kling's text-to-video service - 根目录 > Video Generation > Kling > Official format [Image2video (Image-to-Video Official API)](https://doc-en.302.ai/305983165e0.md): From Kling official API - 根目录 > Video Generation > Kling > Official format [Image2video (Image-to-Video Get Task Result)](https://doc-en.302.ai/305983166e0.md): From Kling's image-to-video service - 根目录 > Video Generation > Kling > Official format [MultiImage2Video（Multiple Images Reference）](https://doc-en.302.ai/320673199e0.md): Multi-image reference from Kling is supported, with a maximum of 4 images allowed. - 根目录 > Video Generation > Kling > Official format [MultiImage2Video（Get Multi-Image Video Task Result）](https://doc-en.302.ai/320673200e0.md): Official website: https://klingai.kuaishou.com/ - 根目录 > Video Generation > Kling > Official format [Effects（Video Effects Official API）](https://doc-en.302.ai/320673201e0.md): From Kling Official API - 根目录 > Video Generation > Kling > Official format [Effects（Get Video Effects Task Result）](https://doc-en.302.ai/320673202e0.md): Official website: https://klingai.kuaishou.com/ - 根目录 > Video Generation > Kling > Official format [Avatar（Digital Human Task Results）](https://doc-en.302.ai/366015340e0.md): The underlying model is kling-v1-6 and cannot be changed. - 根目录 > Video Generation > Kling > Official format [Avatar（Results of the Digital Human Task）](https://doc-en.302.ai/366015341e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Kling > Official format [Identify-Face（Lip-sync Identify face）](https://doc-en.302.ai/382960301e0.md): [Lip-Sync] Identify-Face - 根目录 > Video Generation > Kling > Official format [Advanced-Lip-Sync（Lip-sync Create Task）](https://doc-en.302.ai/382960302e0.md): [Lip Sync] Create Task - 根目录 > Video Generation > Kling > Official format [Advanced-Lip-Sync（Lip Sync Query Task）](https://doc-en.302.ai/382960303e0.md): [Lip Sync] Query Task - 根目录 > Video Generation > CogVideoX [Generations (text-generated video)](https://doc-en.302.ai/261643423e0.md): Video Generation Model from Zhipu - 根目录 > Video Generation > CogVideoX [Generations(Image-generated video)](https://doc-en.302.ai/261643424e0.md): Image-to-Video Model from Zhipu - 根目录 > Video Generation > CogVideoX [Generations (Video Generation from Start and End Frames)](https://doc-en.302.ai/328166283e0.md): Video Generation Model from Zhipu - 根目录 > Video Generation > CogVideoX [Results (get task results)](https://doc-en.302.ai/261643425e0.md): Video Generation Model from Zhipu - 根目录 > Video Generation > Minimax [Video（Text-to-Video）](https://doc-en.302.ai/267624009e0.md): Video generation model from Minimax, supported models: - 根目录 > Video Generation > Minimax [Video (Image-to-Video Generation)](https://doc-en.302.ai/310811463e0.md): Video generation model from Minimax, supported models: - 根目录 > Video Generation > Minimax [Video（Based on Subject Reference）](https://doc-en.302.ai/256735121e0.md): **Specification** - 根目录 > Video Generation > Minimax [Video (Camera Movement Control)](https://doc-en.302.ai/310811464e0.md): When the model parameter is set to T2V-01-Director or I2V-01-Director, the response to camera movement control in the prompt is more accurate. - 根目录 > Video Generation > Minimax [Video（MiniMax-Hailuo-02）](https://doc-en.302.ai/222607630e0.md): The latest video generation model from Minimax, MiniMax-Hailuo-02, supports a variety of parameter combinations to flexibly meet different video generation needs. - 根目录 > Video Generation > Minimax [Video（MiniMax-Hailuo-2.3）](https://doc-en.302.ai/367897234e0.md): The latest video generation models from **Minimax**, **MiniMax-Hailuo-2.3** and **MiniMax-Hailuo-2.3-fast**, support various parameter combinations to flexibly meet different video generation needs. - 根目录 > Video Generation > Minimax [Query（Result）](https://doc-en.302.ai/212308282e0.md): Text-to-video from from Minimax - 根目录 > Video Generation > Minimax [Files（Video Download）](https://doc-en.302.ai/212308283e0.md): Text-to-video from from Minimax - 根目录 > Video Generation > Pika [1.5 pikaffects（Image-to-Video Generation）](https://doc-en.302.ai/285339580e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [Turbo Generate（Text-to-Video Generation）](https://doc-en.302.ai/285339686e0.md): Powered by Pika's video generation model. - 根目录 > Video Generation > Pika [Turbo Generate（Text-to-Video Generation）](https://doc-en.302.ai/285339795e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [2.1 Generate（Text-to-Video Generation）](https://doc-en.302.ai/285339947e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [2.1 Generate（Image-to-Video Generation） ](https://doc-en.302.ai/285340084e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [2.2 Generate（Text-to-Video Generation）](https://doc-en.302.ai/285340208e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [2.2 Generate（Image-to-Video Generation） ](https://doc-en.302.ai/285340561e0.md): Powered by Pika's video generation model. - 根目录 > Video Generation > Pika [2.2 Pikascenes（Generate scene videos） ](https://doc-en.302.ai/285341167e0.md): Powered by Pika's video generation model - 根目录 > Video Generation > Pika [Fetch（Result）](https://doc-en.302.ai/225990814e0.md): Get the video generation task results - 根目录 > Video Generation > PixVerse [Generate（Text-to-Video Generation）](https://doc-en.302.ai/236444989e0.md): Video generation model from PixVerse, generates videos from text, now supports v5.5 - 根目录 > Video Generation > PixVerse [Generate（Image-to-Video Generation）](https://doc-en.302.ai/319453529e0.md): Video generation model from PixVerse, generates videos from images, now supports v5.5 - 根目录 > Video Generation > PixVerse [Generate（Multi-Subject Reference）](https://doc-en.302.ai/319453530e0.md): Video generation model from PixVerse, multi-subject reference - 根目录 > Video Generation > PixVerse [Fetch](https://doc-en.302.ai/236444990e0.md): Retrieve the results of the video generation task. - 根目录 > Video Generation > PixVerse [Lipsync (Submit lip-sync task)](https://doc-en.302.ai/331356568e0.md): Generate realistic lip-sync animations from audio using PixVerse Lipsync model's advanced synchronization algorithm - 根目录 > Video Generation > PixVerse [Lipsync (Get lip-sync task result)](https://doc-en.302.ai/331356569e0.md): Get Task Result - 根目录 > Video Generation > Genmo [Mochi-v1 (Get task results)](https://doc-en.302.ai/256081480e0.md): Video generation model from Tencent open source - 根目录 > Video Generation > Genmo [Mochi-v1（Text to Video）](https://doc-en.302.ai/226300102e0.md): Genmo has set a new benchmark in video generation by open-sourcing Mochi1, their latest model. Featuring an innovative Asymmetric Diffusion Transformer (AsymmDiT) architecture and up to 10 billion parameters, Mochi1 stands as the largest publicly released video generation model to date. - 根目录 > Video Generation > Hedra > 2.0 [Audio（Upload）](https://doc-en.302.ai/226300103e0.md): Audio upload interface from Hedra. - 根目录 > Video Generation > Hedra > 2.0 [Portrait（Upload） ](https://doc-en.302.ai/226300104e0.md): Image upload interface from Hedra. - 根目录 > Video Generation > Hedra > 2.0 [Characters（lip-synthesis）](https://doc-en.302.ai/226300106e0.md): A lip-synthesis interface from Hedra. - 根目录 > Video Generation > Hedra > 2.0 [Fetch（Result）](https://doc-en.302.ai/226300105e0.md): Get video generation task results - 根目录 > Video Generation > Hedra > 3.0 [List Models（Get a list of models）](https://doc-en.302.ai/303478966e0.md): Get a list of models - 根目录 > Video Generation > Hedra > 3.0 [Create Asset（Resource creation） ](https://doc-en.302.ai/303478967e0.md): Resource creation interface from Hedra. - 根目录 > Video Generation > Hedra > 3.0 [Upload Asset（Resource upload） ](https://doc-en.302.ai/303478968e0.md): Resource creation interface from Hedra. - 根目录 > Video Generation > Hedra > 3.0 [Generate Asset（Resource Synthesis）](https://doc-en.302.ai/303478969e0.md): Resource creation interface from Hedra. - 根目录 > Video Generation > Hedra > 3.0 [Get Status（Get resource synthesis results）](https://doc-en.302.ai/303478970e0.md): Get a list of models - 根目录 > Video Generation > Haiper [Haiper（Text to Video）](https://doc-en.302.ai/232137062e0.md): A rapidly growing video generation company based in London - 根目录 > Video Generation > Haiper [Haiper（Image to Video）](https://doc-en.302.ai/232137063e0.md): A rapidly growing video generation company based in London - 根目录 > Video Generation > Haiper [Haiper（Text to Video V2.5）](https://doc-en.302.ai/263666434e0.md): An emerging video generation company from London - 根目录 > Video Generation > Haiper [Haiper（Image to Video V2.5）](https://doc-en.302.ai/263666435e0.md): An emerging video generation company from London - 根目录 > Video Generation > Haiper [Haiper（Fetch Task Result）](https://doc-en.302.ai/252700108e0.md): A video generation model open-sourced by Tencent - 根目录 > Video Generation > Sync. [Generate](https://doc-en.302.ai/234772368e0.md): Upload a video and audio to perform lip-sync matching, from: https://sync.so/ - 根目录 > Video Generation > Sync. [Fetch](https://doc-en.302.ai/234772369e0.md): Get Video Generation Task Result - 根目录 > Video Generation > Lightricks [Ltx-Video](https://doc-en.302.ai/237132298e0.md): Open-source video model, characterized by extremely fast generation speed. - 根目录 > Video Generation > Lightricks [Ltx-Video-I2V](https://doc-en.302.ai/237132299e0.md): Open-source video model, characterized by extremely fast generation speed. - 根目录 > Video Generation > Lightricks [Ltx-Video-v095（Text-to-video generation）](https://doc-en.302.ai/268482790e0.md): Open-source video model characterized by fast generation speed. - 根目录 > Video Generation > Lightricks [Ltx-Video-v095-I2V（Image-to-Video Generation）](https://doc-en.302.ai/268489487e0.md): Open-source video model characterized by fast generation speed. - 根目录 > Video Generation > Hunyuan [Hunyuan（Text-to-Video）](https://doc-en.302.ai/241433337e0.md): Video generation model from Tencent Open Source - 根目录 > Video Generation > Hunyuan [Hunyuan（Obtain Task Results） ](https://doc-en.302.ai/241433338e0.md): Video generation model from Tencent Open Source - 根目录 > Video Generation > Vidu [Vidu（Text-to-Video） ](https://doc-en.302.ai/244703992e0.md): A rising video generation company from China, whose latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Vidu（Image to Video） ](https://doc-en.302.ai/244703993e0.md): A new emerging video generation company in China has introduced its latest 1.5 model, featuring an exclusive subject generation capability - 根目录 > Video Generation > Vidu [Vidu（Generate video from the first and last frames） ](https://doc-en.302.ai/244703994e0.md): A new emerging video generation company in China has introduced its latest 1.5 model, featuring an exclusive subject generation capability - 根目录 > Video Generation > Vidu [Vidu（Reference-based video generation） ](https://doc-en.302.ai/244703995e0.md): A rising video generation company from China, whose latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Vidu（Generate scene video） ](https://doc-en.302.ai/244703996e0.md): A rising video generation company from China. - 根目录 > Video Generation > Vidu [Vidu（Smart Ultra HD） ](https://doc-en.302.ai/244703997e0.md): A rising video generation company from China, whose latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Fetch（Retrieve Task Results）](https://doc-en.302.ai/244703998e0.md): Retrieve Video Generation Task Results - 根目录 > Video Generation > Vidu [Vidu V2（Text-to-Video Generation） ](https://doc-en.302.ai/294939973e0.md): A rising domestic video generation company. Its latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Vidu V2（Image-to-Video Generation） ](https://doc-en.302.ai/294940013e0.md): An emerging domestic video generation company; the 1.5 model offers an exclusive subject generation feature - 根目录 > Video Generation > Vidu [Vidu V2（First-and-Last-Frame Video Generation） ](https://doc-en.302.ai/294940083e0.md): An emerging China-based video generation company; its latest 1.5 model offers an exclusive subject-generation feature - 根目录 > Video Generation > Vidu [Vidu V2（Subject-Driven Video Generation） ](https://doc-en.302.ai/294940163e0.md): A rising domestic video generation company. Its latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Vidu（Scene Video Generation V2） ](https://doc-en.302.ai/287054001e0.md): An emerging domestic company specializing in video generation. - 根目录 > Video Generation > Vidu [Vidu V2（AI Ultra HD – Premium） ](https://doc-en.302.ai/294942159e0.md): A rising domestic video generation company. Its latest 1.5 model features an exclusive subject generation capability. - 根目录 > Video Generation > Vidu [Fetch V2（Retrieve Task Result）](https://doc-en.302.ai/294940231e0.md): Retrieve the result of a video generation task. - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [I2V (Image-to-Video Alibaba Cloud)](https://doc-en.302.ai/331998204e0.md): ## Supported Models - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [T2V (Text-to-Video Alibaba Cloud)](https://doc-en.302.ai/331998203e0.md): Supported Models - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [R2V (Reference-Generated Video)](https://doc-en.302.ai/393391921e0.md): Supported models: - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [wan2.2-animate-move(Image-to-Motion)](https://doc-en.302.ai/354415458e0.md): From Tongyi Wanxiang's wan2.2-animate-move Image-to-Motion Model - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [wan2.2-animate-mix (Video replacement)](https://doc-en.302.ai/354583214e0.md): Wanxiang wan2.2-animate-mix video face-swapping model allows you to replace the main character in a video with the character from a given image, while preserving the original video's scene, lighting, and tone for seamless face-swapping. - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [wan2.2-s2v（Digital Human Generation）](https://doc-en.302.ai/342267514e0.md): The Digital Human wan2.2-s2v model can generate videos with natural movements, speaking, singing, or performing, based on a single image and audio input. - 根目录 > Video Generation > Tongyi Wanxiang > Aliyun [Tasks (Get Task Results)](https://doc-en.302.ai/331998205e0.md): An open-source video generation model from Tencent - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-a14b-t2v (Text to Video)](https://doc-en.302.ai/329405865e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-a14b-t2v (Get Task Result)](https://doc-en.302.ai/329405866e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-5b-t2v (Text to Video)](https://doc-en.302.ai/329405869e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-5b-t2v (Get Task Result)](https://doc-en.302.ai/329405870e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-t2v（Text-to-video open source version）](https://doc-en.302.ai/265218437e0.md): The latest open-source video generation model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-t2v（Fetch Task Result） ](https://doc-en.302.ai/265218438e0.md): The latest open-source video generation model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-vace（Fetch Task Result）](https://doc-en.302.ai/301663560e0.md): The latest open source video editing model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-vace（Video Edit）](https://doc-en.302.ai/301645905e0.md): The latest open source video editing model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [Tasks（Fetch Task Result） ](https://doc-en.302.ai/254095111e0.md): **Price：0 PTC/call** - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-a14b-i2v (Image to Video)](https://doc-en.302.ai/329405867e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-a14b-i2v (Get Task Result)](https://doc-en.302.ai/329405868e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-i2v（Image-to-video open source version）](https://doc-en.302.ai/265218439e0.md): The latest open-source video generation model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan-i2v（Fetch Task Result） ](https://doc-en.302.ai/265218440e0.md): The latest open-source video generation model from Alibaba - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-5b-i2v (Image to Video)](https://doc-en.302.ai/329405871e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Tongyi Wanxiang > Open Source [wan2.2-5b-i2v (Get Task Result)](https://doc-en.302.ai/329405872e0.md): From Alibaba's latest open-source video generation model - 根目录 > Video Generation > Jimeng [Seaweed (Text/picture generated video)](https://doc-en.302.ai/256341887e0.md): Video generation model from Jimeng, supports Wensheng video and Tusheng video (Tusheng video only supports up to 5s) - 根目录 > Video Generation > Jimeng [Seaweed (Fetch Task Results)](https://doc-en.302.ai/256341888e0.md): **Price: 0 PTC/call** - 根目录 > Video Generation > Jimeng [Seedance (Text/picture generated video)](https://doc-en.302.ai/305958132e0.md): ### Interface Parameter Input Reference - 根目录 > Video Generation > Jimeng [Seedance (Video Generation from First and Last Frames)](https://doc-en.302.ai/344157192e0.md): ### **Only the following models are supported currently: - 根目录 > Video Generation > Jimeng [Seedance (Reference-based Video Generation)](https://doc-en.302.ai/344157193e0.md): **Only doubao-seedance-1-0-lite-i2v-250428 supports reference video generation** - 根目录 > Video Generation > Jimeng [Seedance (Fetch Task Results)](https://doc-en.302.ai/305958133e0.md): **Price: 0 PTC/call** - 根目录 > Video Generation > Jimeng [Omnihuman-1（Submits Task）](https://doc-en.302.ai/330039598e0.md): Generate a video effect corresponding to the uploaded image based on a single image and audio provided by the user. The OmniHuman digital human model supports input of any aspect ratio image containing a person, and, combined with audio, generates vivid and high-quality video results. The character's emotions and movements are highly correlated with the audio. At the same time, the model has a certain generalization ability for anime, cartoon, and other styles, making it widely applicable in scenarios such as content expression, singing, and performance. - 根目录 > Video Generation > Jimeng [Omnihuman-1（Get Task Results）](https://doc-en.302.ai/330039599e0.md): **Price: Free** - 根目录 > Video Generation > Jimeng [Omnihuman-1.5(Subject detection)](https://doc-en.302.ai/377782504e0.md): Detects the subject in an image and supports returning the corresponding mask image. This step can be skipped if the subject doesn't need to speak during video generation. - 根目录 > Video Generation > Jimeng [Omnihuman-1.5(Submit Task)](https://doc-en.302.ai/377782505e0.md): OmniHuman 1.5 (also known as the Jimeng Source Digital Human Model) generates video effects corresponding to user-uploaded single images and audio. It supports inputting images of any frame containing people or other subjects (pets, anime, etc.), combining them with audio to generate high-quality videos. - 根目录 > Video Generation > Jimeng [Omnihuman-1.5(Query)](https://doc-en.302.ai/377782506e0.md): **Price：0PTC** - 根目录 > Video Generation > SiliconFlow [Video (Video Generation) ](https://doc-en.302.ai/255033894e0.md): Open-source video models from SiliconFlow. - 根目录 > Video Generation > SiliconFlow [Tasks(Fetch Task Result) ](https://doc-en.302.ai/255033897e0.md): Fetch Task Result - 根目录 > Video Generation > Google [Veo3.1（Text-to-Video）](https://doc-en.302.ai/361708990e0.md): Google's Latest Veo3.1 Version - 根目录 > Video Generation > Google [Veo3.1（Image-to-Video）](https://doc-en.302.ai/361940345e0.md): Google's latest Veo3.1 version - 根目录 > Video Generation > Google [Veo3.1（Get results）](https://doc-en.302.ai/361708991e0.md): **Price: Free** - 根目录 > Video Generation > Google [Veo3-Fast（Text-to-video）](https://doc-en.302.ai/314500532e0.md): Latest video generation model from Google, generates 8s videos with sound effects - 根目录 > Video Generation > Google [Veo3-Fast（Get task result）](https://doc-en.302.ai/314500533e0.md): **Price: 0 PTC/call** - 根目录 > Video Generation > Google [Veo3-Fast-Frames（Image and Text to Video Generation）](https://doc-en.302.ai/323639969e0.md): The latest video generation model from Google supports input images and can generate 8-second videos with sound effects at a time. - 根目录 > Video Generation > Google [Veo3-Fast-Frames (Get task result)](https://doc-en.302.ai/323639970e0.md): **Price: Free** - 根目录 > Video Generation > Google [Veo3-Pro（Text-to-video）](https://doc-en.302.ai/314500534e0.md): Latest video generation model from Google, generates 8s videos with sound effects - 根目录 > Video Generation > Google [Veo3-Pro（Get task result）](https://doc-en.302.ai/314500535e0.md): **Price: 0 PTC/call** - 根目录 > Video Generation > Google [Veo3-Pro-Frames（Image and Text to Video Generation）](https://doc-en.302.ai/315837722e0.md): Google's latest Veo3-Pro-Frames video generation model, generates 8-second videos with sound effects in a single task - 根目录 > Video Generation > Google [Veo3-Pro-Frames（Veo3-Pro-Frames）](https://doc-en.302.ai/315837723e0.md): **Price: Free** - 根目录 > Video Generation > Google [Veo2（Text-to-video）](https://doc-en.302.ai/263672348e0.md): The latest video generation model from Google - 根目录 > Video Generation > Google [Veo2-i2v（Image to video generation）](https://doc-en.302.ai/305943205e0.md): The latest video generation model from Google - 根目录 > Video Generation > Google [Veo2（Get task results） ](https://doc-en.302.ai/263672349e0.md): The latest video generation model from Google - 根目录 > Video Generation > Google [Veo3 (Text-to-video)](https://doc-en.302.ai/305943206e0.md): From Google's latest video generation model, generating 8s videos with sound effects in one go - 根目录 > Video Generation > Google [Veo3 (Get task result)](https://doc-en.302.ai/305943207e0.md): From Google's latest video generation model - 根目录 > Video Generation > Google [Veo3-V2（V2 API Format）](https://doc-en.302.ai/332397149e0.md): The V2 version API format can call the Veo3 series models, and charges will be applied according to the corresponding model. - 根目录 > Video Generation > Google [Get Task（V2 API Format）](https://doc-en.302.ai/332397150e0.md): **Price: Free** - 根目录 > Video Generation > Kunlun Tech [Skyreels（Image to Video）](https://doc-en.302.ai/263672450e0.md): The latest video generation model from Kunlun Tech - 根目录 > Video Generation > Kunlun Tech [Skyreels（Get task results） ](https://doc-en.302.ai/263672451e0.md): The latest video generation model from Kunlun Tech - 根目录 > Video Generation > Higgsfield > Offical Format [ Motions (Get Template List)](https://doc-en.302.ai/346048829e0.md): **Price: Free / per use** - 根目录 > Video Generation > Higgsfield > Offical Format [Generate (Official Image-to-Video Generation)](https://doc-en.302.ai/346048830e0.md): A video generation model from higgsfield.ai that generates videos from images. - 根目录 > Video Generation > Higgsfield > Offical Format [Speak (Digital Human Generation)](https://doc-en.302.ai/346048831e0.md): Digital human generation from Higgsfield. Input an image and audio to generate a digital human. The maximum duration is 15 seconds. - 根目录 > Video Generation > Higgsfield > Offical Format [Fetch (Get Task Results)](https://doc-en.302.ai/346048832e0.md): Get Image Generation Task or Feature Generation Task Result - 根目录 > Video Generation > Higgsfield [Generate（Image to Video） ](https://doc-en.302.ai/301920549e0.md): - 根目录 > Video Generation > Higgsfield [Shortads（Image-Generated Advertising Video） ](https://doc-en.302.ai/299579987e0.md): The advertising video generation workflow from higgsfield.ai - 根目录 > Video Generation > Higgsfield [Apps (Image-to-Video)](https://doc-en.302.ai/347477581e0.md): Automatically edit images and generate videos based on predefined templates - 根目录 > Video Generation > Higgsfield [Fetch (Get Task Results)](https://doc-en.302.ai/294544225e0.md): Get Image Generation Task or Feature Generation Task Result - 根目录 > Video Generation > Chanjing [Create a video synthesis task](https://doc-en.302.ai/312102536e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/ydxXB52LJq7yOZ7zuaY3Rv0NWqjMp697 - 根目录 > Video Generation > Chanjing [Retrieve video details](https://doc-en.302.ai/312102537e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/Obva6QBXJw9yve90tZNjvPzNWn4qY5Pr - 根目录 > Video Generation > Chanjing [Delete video](https://doc-en.302.ai/312102538e0.md): Asynchronous task - 根目录 > Video Generation > Chanjing [Retrieve the list of supported fonts](https://doc-en.302.ai/312102539e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/kDnRL6jAJM3yQX3zH6RPd1GGWyMoPYe1 - 根目录 > Video Generation > Chanjing [Generate a digital human avatar](https://doc-en.302.ai/312102540e0.md): Asynchronous task, specific progress can be checked through the query interface - 根目录 > Video Generation > Chanjing [Retrieve avatar details](https://doc-en.302.ai/312102541e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/m9bN7RYPWdlQDxlKhvQPqDObWZd1wyK0 - 根目录 > Video Generation > Chanjing [Delete avatar](https://doc-en.302.ai/312102542e0.md): - 根目录 > Video Generation > Chanjing [Public digital human list](https://doc-en.302.ai/312102543e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/lyQod3RxJK3xYz3pHnQZ3l0wJkb4Mw9r - 根目录 > Video Generation > Midjourney [MJ-Video（Image to Video）](https://doc-en.302.ai/315837724e0.md): Image-to-Video Model from Midjourney, each task generates 4 videos, each video 5 seconds - 根目录 > Video Generation > Midjourney [MJ-Video（Video Extension）](https://doc-en.302.ai/315837725e0.md): Image-to-Video Model from Midjourney, supports extending generated videos by 5 seconds - 根目录 > Video Generation > Midjourney [Fetch（Fetch Task）](https://doc-en.302.ai/315837726e0.md): **Price：0 PTC/call** - 根目录 > Video Generation > Topview > Marketing Digital Avatar [Submit Avatar Marketing Video](https://doc-en.302.ai/321992012e0.md): Avatar Marketing Video is Topview's product marketing video generation tool. Users can upload product images, scripts, or materials to easily generate videos demonstrating the product being worn or interacted with, supporting multiple languages. - 根目录 > Video Generation > Topview > Marketing Digital Avatar [Get Avatar Marketing Video Results](https://doc-en.302.ai/321992013e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Marketing Digital Avatar [Get Script List](https://doc-en.302.ai/321992014e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Marketing Digital Avatar [Modify Script Content](https://doc-en.302.ai/321992015e0.md): After the update, call this interface to query the Avatar Marketing Video result - 根目录 > Video Generation > Topview > Regular Digital Avatar [VideoAvatar Submit](https://doc-en.302.ai/321992016e0.md): VideoAvatar is an AI digital human video generation interface provided by the Topview platform. It allows users to easily generate dynamic videos featuring digital human explanations by selecting an AI Avatar and audio material. This capability is widely used in content creation, marketing videos, and more. - 根目录 > Video Generation > Topview > Regular Digital Avatar [VideoAvatar Query](https://doc-en.302.ai/321992017e0.md): The generation speed is relatively slow, and it may take a while for the video to appear after the link is returned. - 根目录 > Video Generation > Topview > Regular Digital Avatar [Create Private Digital Avatar](https://doc-en.302.ai/321992018e0.md): Video2aiAvatar is the core API of the Topview platform for converting local videos into exclusive AI digital human (AI Avatar) models with one click. Before using this API, you need to upload your local video via the file upload API and obtain the fileId. - 根目录 > Video Generation > Topview > Regular Digital Avatar [Query Private Digital Avatar](https://doc-en.302.ai/321992019e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Regular Digital Avatar [Delete Private Digital Avatar](https://doc-en.302.ai/321992020e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Regular Digital Avatar [Query Public Digital Avatar](https://doc-en.302.ai/321992021e0.md): Query public digital humans on the Topview platform - 根目录 > Video Generation > Topview > Regular Digital Avatar [Query Public Voices](https://doc-en.302.ai/321992022e0.md): This API is used to query all available public Voice resources within the Topview platform. - 根目录 > Video Generation > Topview > Regular Digital Avatar [Query Subtitle Style Interface](https://doc-en.302.ai/321992023e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Digital Avatar [Product ImageReplace Submit](https://doc-en.302.ai/321992024e0.md): Product ImageReplace is an intelligent product image replacement API for digital human images provided by the Topview platform. It allows users to replace specified product images with template images, automatically generating new digital human product images. This capability is widely used in e-commerce display, marketing content, and other scenarios. - 根目录 > Video Generation > Topview > Product Digital Avatar [Product ImageReplace Query](https://doc-en.302.ai/321992025e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Digital Avatar [Product Image2Video Submit](https://doc-en.302.ai/321992026e0.md): Product Image2Video is an API provided by the Topview platform that enables the conversion of product digital human images into videos. It allows users to automatically generate dynamic videos from specified product images, digital human figures, and other materials using AI technology. This capability is widely used in scenarios such as e-commerce display, marketing content, and short video creation. - 根目录 > Video Generation > Topview > Product Digital Avatar [Product Image2Video Query](https://doc-en.302.ai/321992027e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Digital Avatar [Query Public Digital Avatar](https://doc-en.302.ai/321992028e0.md): This API is used to query all available public digital human avatars in the Topview platform. - 根目录 > Video Generation > Topview > Product Digital Avatar [Query Product Categories](https://doc-en.302.ai/321992029e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Digital Avatar [Product Image2Video V2 Summit](https://doc-en.302.ai/388945276e0.md): Product Image2Video is an API for converting product digital human image to video provided by the Topview platform. It supports users to automatically generate dynamic videos from specified materials such as product images and digital human images through AI technology, and is widely used in scenarios such as e-commerce display, marketing content, and short video creation. - 根目录 > Video Generation > Topview > Product Digital Avatar [Product Image2Video V2 Query](https://doc-en.302.ai/388945277e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Image Replacement [productAnyShoot ReplaceImage Submit](https://doc-en.302.ai/321992030e0.md): productAnyShoot ReplaceImage is an intelligent product image replacement API provided by the Topview platform. It allows users to replace a specified product image with a template image and automatically generate a new product display image. - 根目录 > Video Generation > Topview > Product Image Replacement [productAnyShoot ReplaceImage Query](https://doc-en.302.ai/321992031e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Image Replacement [productAnyShoot ReplaceImage V3 Summit](https://doc-en.302.ai/391749014e0.md): Use product images, product mask images, model images, and model mask images to create composite result images. - 根目录 > Video Generation > Topview > Product Image Replacement [productAnyShoot ReplaceImage V3 Query](https://doc-en.302.ai/391751529e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Image Replacement [Product ImageReplace V3 Submission](https://doc-en.302.ai/388945278e0.md): Product ImageReplace is an interface for intelligent replacement of product digital human image commodity pictures provided by the Topview platform. It supports users to replace specified commodity pictures with template pictures and automatically generate new product digital human images. It is widely used in scenarios such as e-commerce display and marketing content. - 根目录 > Video Generation > Topview > Product Image Replacement [Product ImageReplace V3 Query](https://doc-en.302.ai/388945279e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Product Image Replacement [Remove Background Generate product images without background Submit](https://doc-en.302.ai/388945280e0.md): Remove Background is a feature provided by the Topview platform to generate product images without a background. - 根目录 > Video Generation > Topview > Product Image Replacement [Remove Background Generate product image query without background](https://doc-en.302.ai/388945281e0.md): **Price: 0 PTC/Time** - 根目录 > Video Generation > Topview > Product Image Replacement [Query Template List](https://doc-en.302.ai/321992032e0.md): Used to obtain a list of all available product smart shooting templates - 根目录 > Video Generation > Topview > Product Image Replacement [Query Template Categories](https://doc-en.302.ai/321992033e0.md): Used to obtain category information for all product shooting templates - 根目录 > Video Generation > Topview > Image to Video [Submit Image2video (Image to Video)](https://doc-en.302.ai/321992034e0.md): Image2Video is an AI-powered image-to-video tool launched by Topview. Users only need to upload an image, and within minutes, static images can be transformed into dynamic videos. - 根目录 > Video Generation > Topview > Image to Video [Query Image2video (Image to Video)](https://doc-en.302.ai/321992035e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Submit Digital Human Generation Task](https://doc-en.302.ai/347477582e0.md): Reference: https://docs.topview.ai/reference/get_new-endpoint-58 - 根目录 > Video Generation > Topview > Avatar 4 [Get Digital Human Generation Task Result](https://doc-en.302.ai/347477583e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Query Subtitle Style API](https://doc-en.302.ai/347477584e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Query Public Voice Tones](https://doc-en.302.ai/347477585e0.md): This API is used to query all available public Voice resources within the Topview platform. - 根目录 > Video Generation > Topview > Avatar 4 [Query Available Digital Human List](https://doc-en.302.ai/347477586e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Query Digital Human Category List](https://doc-en.302.ai/347477587e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Create Custom Digital Human](https://doc-en.302.ai/347477588e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Delete Custom Digital Human](https://doc-en.302.ai/347477589e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Create Text-to-Speech Task](https://doc-en.302.ai/347477590e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview > Avatar 4 [Query Text-to-Speech Task](https://doc-en.302.ai/347477591e0.md): **Price: 0 PTC/time** - 根目录 > Video Generation > Topview [Upload Interface](https://doc-en.302.ai/321992011e0.md): Currently, only mp4, mov, png, jpg, jpeg, bmp, webp, mp3, wav, and m4a are supported. - 根目录 > Video Generation > OpenAI [Chat（Video generation）](https://doc-en.302.ai/357839019e0.md): The latest video generation model from OpenAI, Sora-2, can be integrated into any Chatbot using the v1/chat/completions format API. - 根目录 > Video Generation > OpenAI [Sora2（Asynchronous request）](https://doc-en.302.ai/357839020e0.md): Latest video generation model from OpenAI: Sora-2 - 根目录 > Video Generation > OpenAI [Sora2（Get task result） ](https://doc-en.302.ai/357839021e0.md): **Price: Free** - 根目录 > Video Generation > OpenAI [Create Video（Official Format）](https://doc-en.302.ai/359167324e0.md): **Sora** is OpenAI's latest video generation model, capable of generating detailed and dynamic videos based on text or images. - 根目录 > Video Generation > OpenAI [Retrieve Video（Official Format）](https://doc-en.302.ai/359167325e0.md): Enter the task ID to check the generation progress - 根目录 > Video Generation > OpenAI [Remix Video（Official Format）](https://doc-en.302.ai/359167326e0.md): sora2 supports Remix video, allowing secondary creation based on an existing video ID and new prompt, enabling "existing video + new intent" regeneration. - 根目录 > Video Generation > OpenAI [Delete Video（Official Format）](https://doc-en.302.ai/359167327e0.md): Delete a generated video by video_id - 根目录 > Video Generation > OpenAI [Retrieve Content（Official Format）](https://doc-en.302.ai/359167328e0.md): **Price: 0 PTC** - 根目录 > Video Generation > Wavespeed > Universal Interface [Wavespeed Generation Task](https://doc-en.302.ai/359497924e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in pricing. - 根目录 > Video Generation > Wavespeed > Universal Interface [Wavespeed Retrieval Task](https://doc-en.302.ai/359497925e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.6/text-to-video](https://doc-en.302.ai/392233003e0.md): Alibaba's Wanxiang 2.6 model can convert plain text prompts (with optional audio) into 5-15 second movie clips. It supports multi-shot narrative, vertical or horizontal format, with a resolution of up to 1080p, making it ideal for advertisements, trailers, and social media content. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.6/image-to-video](https://doc-en.302.ai/392233004e0.md): alibaba/wan-2.6/image-to-video can convert text or images into videos with Audio And Video Synchronization (supporting 720p/1080p resolution). Compared to Google Veo3, it has a faster conversion speed and lower cost, and can transform a single image + text prompt into a 5–15 second high-quality dynamic video, supporting synchronized audio (reserved field) and multi-shot storytelling. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.6/reference-to-video](https://doc-en.302.ai/392233005e0.md): alibaba/wan-2.6/reference-to-video can convert character, prop, or scene references (single-view or multi-view) into new video shots while preserving their characteristics, style, and layout, and achieving smooth and coherent motion. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.5/text-to-video-fast](https://doc-en.302.ai/360390004e0.md): Alibaba WAN 2.5 is a state-of-the-art text/image-to-video generation model available on Alibaba Cloud's DashScope platform. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.2/t2v-plus-1080p](https://doc-en.302.ai/362598483e0.md): The Pika V2.0 Turbo text-to-video model can generate high-quality videos based on text descriptions, supports multiple video resolutions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [alibaba/wan-2.5/image-to-video](https://doc-en.302.ai/363568415e0.md): Alibaba WAN 2.5 is the most advanced text/image-to-video generation model on Alibaba Cloud DashScope platform. This powerful model can generate high-quality videos from simple text or image prompts, supporting 480p, 720p, or 1080p resolution, and comes with synchronized audio. Compared to Google Veo3, it offers a more affordable solution and faster video generation speed, making it an ideal choice for efficiently creating videos with embedded audio. - 根目录 > Video Generation > Wavespeed [bytedance/seedance-v1-lite-t2v-720p](https://doc-en.302.ai/360850245e0.md): ByteDance Seedance is an advanced AI video generation model that excels at producing coherent multi-shot videos. - 根目录 > Video Generation > Wavespeed [bytedance/seedance-v1-pro-t2v-720p](https://doc-en.302.ai/360418207e0.md): ByteDance Seedance is an advanced AI video generation model that excels at creating coherent multi-shot videos. - 根目录 > Video Generation > Wavespeed [bytedance/seedance-v1-lite-t2v-1080p](https://doc-en.302.ai/361512271e0.md): ByteDance's Seedance is an advanced AI video generation model that excels at producing coherent multi-shot videos. - 根目录 > Video Generation > Wavespeed [bytedance/dreamina-v3.0-pro/text-to-video](https://doc-en.302.ai/361597676e0.md): Dreamina V3.0 Pro features enhanced text and image comprehension capabilities, enabling the generation of 1080P high-definition videos with more precise and natural dynamic performance, delivering professional-grade visual quality. - 根目录 > Video Generation > Wavespeed [bytedance/dreamina-v3.0/text-to-video-1080p](https://doc-en.302.ai/362762476e0.md): Dreamina V3.0 offers more precise and naturally dynamic responses to text or image inputs, capable of generating high-quality 1080P videos with professional-level visual effects. This version supports diverse and expressive styles, and presents multiple scenes within narratives. - 根目录 > Video Generation > Wavespeed [bytedance/seedance-v1-pro-t2v-1080p](https://doc-en.302.ai/360624800e0.md): ByteDance Seedance is an advanced AI video generation model that excels at producing coherent multi-shot videos. - 根目录 > Video Generation > Wavespeed [bytedance/dreamina-v3.0/text-to-video-720p](https://doc-en.302.ai/362162068e0.md): Dreamina V3.0 features higher response accuracy and natural dynamic performance, capable of generating 720P high-quality videos based on text or image input, delivering professional-level visual effects. - 根目录 > Video Generation > Wavespeed [bytedance/video-upscaler](https://doc-en.302.ai/373987935e0.md): Enhance and upscale your videos with ByteDance's AI video super-resolution technology. In a secure cloud environment, advanced AI technology can instantly improve image clarity, restore fine details, and deliver stunning 4K visual effects. - 根目录 > Video Generation > Wavespeed [character-ai/ovi/image-to-video](https://doc-en.302.ai/363568417e0.md): InfiniteTalk is an audio-driven conversational AI video generation model. With a single image and audio input, you can create dialogue or singing videos. Our pricing is $0.15 per 5 seconds of video generation (480p) or $0.3 (720p), and we support video generation up to 10 minutes in length. - 根目录 > Video Generation > Wavespeed [character-ai/ovi/text-to-video](https://doc-en.302.ai/359725866e0.md): Ovi is a video+audio generation model similar to veo-3, capable of generating video and audio content simultaneously based on text or text+image input. - 根目录 > Video Generation > Wavespeed [google/veo3](https://doc-en.302.ai/360624801e0.md): Google flagship Veo 3 text-to-video generation model—now supports audio generation. - 根目录 > Video Generation > Wavespeed [google/veo3.1/text-to-video](https://doc-en.302.ai/361757950e0.md): Google Veo 3.1 introduces native 1080p resolution, providing creators with higher image quality and greater creative flexibility. - 根目录 > Video Generation > Wavespeed [google/veo3.1/image-to-video](https://doc-en.302.ai/361757951e0.md): Google Veo 3.1 introduces native 1080p resolution, providing creators with higher image quality and greater flexibility. - 根目录 > Video Generation > Wavespeed [google/veo3.1-fast/text-to-video](https://doc-en.302.ai/361757952e0.md): Google Veo 3.1 introduces native 1080p resolution, providing creators with higher image quality and greater creative flexibility. - 根目录 > Video Generation > Wavespeed [google/veo3.1-fast/image-to-video](https://doc-en.302.ai/361757953e0.md): Google Veo 3.1 quickly introduced native 1080p resolution, bringing creators higher image quality and greater flexibility. - 根目录 > Video Generation > Wavespeed [google/veo3.1/reference-to-video](https://doc-en.302.ai/361757954e0.md): Generate a video in which the specific subject remains consistent with the provided reference image. - 根目录 > Video Generation > Wavespeed [google/veo3-fast](https://doc-en.302.ai/360418204e0.md): Generate videos using Google Veo 3 Fast — faster and more cost-effective compared to the standard Veo 3. - 根目录 > Video Generation > Wavespeed [google/veo2](https://doc-en.302.ai/362762477e0.md): Dreamina V3.0 offers more precise and naturally dynamic responses to text or image inputs, capable of generating high-quality 1080P videos with professional-level visual effects. This version supports diverse and expressive styles, and presents multiple scenes within narratives. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-v1.6-t2v-standard](https://doc-en.302.ai/361597675e0.md): Kling AI released the latest 1.6 model on December 19, 2024. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-lipsync/text-to-video](https://doc-en.302.ai/362162065e0.md): Kling TextToVideo is an advanced text-to-video generation model that can produce realistic lip movements perfectly synchronized with the input text. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-v2.5-turbo-pro/text-to-video](https://doc-en.302.ai/360418205e0.md): Kling 2.5 Turbo Pro: a cutting-edge text-to-video generation solution featuring unparalleled motion smoothness, cinematic visual effects, and exceptional prompt parsing accuracy. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-v2.1-t2v-master](https://doc-en.302.ai/361512267e0.md): Kling v2.1 is the latest evolution of the AI video generation model launched by Kuaishou, capable of seamlessly generating short videos (typically 5–10 seconds, with resolutions of 720p or 1080p) from a single image or text prompt, achieving cinematic-level visual quality. Compared to Kling 2.0, it represents a significant leap forward, with major improvements in motion realism, visual coherence, and prompt accuracy. - 根目录 > Video Generation > Wavespeed [luma/ray-2-t2v](https://doc-en.302.ai/362162069e0.md): The Luma Ray 2 text-to-video generation model can create high-quality videos based on text descriptions. - 根目录 > Video Generation > Wavespeed [luma/ray-2-flash-t2v](https://doc-en.302.ai/362598480e0.md): Luma Ray 2 Flash text-to-video generation model can create high-quality videos based on textual descriptions, supports multiple video dimensions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [luma/ray-1.6-t2v](https://doc-en.302.ai/362762478e0.md): Luma Ray 1.6 text-to-video generation model can create high-quality videos based on textual descriptions, supports multiple video dimensions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [leonardoai/motion-2.0](https://doc-en.302.ai/368726399e0.md): Leonardo Motion 2.0 is a major upgrade to Leonardo's video generation capabilities. Compared to the previous generation, it can produce more realistic and detailed videos. - 根目录 > Video Generation > Wavespeed [minimax/hailuo-02/i2v-pro](https://doc-en.302.ai/360418203e0.md): Hailuo 02 is a next-generation AI video generation model launched by Hailuo AI, built on the continuously evolving MiniMax framework. It has been finely tuned to output ultra-clear 1080P resolution videos and features unprecedented response speed, even realistically handling the most complex and extreme physics-driven scenarios. - 根目录 > Video Generation > Wavespeed [minimax/hailuo-02/t2v-standard](https://doc-en.302.ai/360418206e0.md): Hailuo 02 is a next-generation AI video generation model launched by Hailuo AI, built on the continuously evolving MiniMax framework. - 根目录 > Video Generation > Wavespeed [vidu/text-to-video](https://doc-en.302.ai/361597677e0.md): The Vidu text-to-video model is capable of generating high-quality 720p videos, featuring exceptional visual effects and a rich variety of motion expressiveness. - 根目录 > Video Generation > Wavespeed [pika/v2.0-turbo-t2v](https://doc-en.302.ai/362598482e0.md): The Pika V2.0 Turbo text-to-video model can generate high-quality videos based on text descriptions, supports multiple video resolutions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [pika/v2.2-t2v](https://doc-en.302.ai/362598478e0.md): Pika V2.2 text-to-video generation model can create high-quality videos based on textual descriptions, supports multiple video dimensions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [pika/v2.1-t2v](https://doc-en.302.ai/362598479e0.md): Pika V2.1 text-to-video generation model can create high-quality videos based on textual descriptions, supports multiple video dimensions, and features advanced prompt optimization capabilities. - 根目录 > Video Generation > Wavespeed [seedance-v1-pro-t2v-480p](https://doc-en.302.ai/359725867e0.md): Seedance by ByteDance is an advanced AI video model that excels at generating coherent multi-shot videos. It delivers smooth and stable motion effects, and can accurately produce detailed video content based on complex prompts. - 根目录 > Video Generation > Wavespeed [seedance-v1-lite-t2v-480p](https://doc-en.302.ai/359725868e0.md): Seedance by ByteDance is an advanced AI video model that excels at generating coherent multi-shot videos. It delivers smooth and stable motion effects, and can accurately produce detailed video content based on complex prompts. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-v2.0-t2v-master](https://doc-en.302.ai/362598484e0.md): Kling 2.0 is the latest generation AI video generation model launched by Kling AI, featuring significant improvements over the previous Kling 1.6 model. This version offers stronger prompt comprehension, smoother and more natural character motion, higher visual quality, and introduces the brand-new Multi-Elements Editor, making the video editing process more intuitive, flexible, and efficient. - 根目录 > Video Generation > Wavespeed [openai/sora-2/text-to-video](https://doc-en.302.ai/359725870e0.md): OpenAI's Sora 2 is a brand-new, state-of-the-art video and audio generation model. - 根目录 > Video Generation > Wavespeed [openai/sora-2/image-to-video](https://doc-en.302.ai/360390001e0.md): OpenAI's Sora 2 is a brand-new, state-of-the-art video and audio generation model. - 根目录 > Video Generation > Wavespeed [openai/sora-2/text-to-video-pro](https://doc-en.302.ai/360390002e0.md): Sora 2 from OpenAI is a brand-new, state-of-the-art video and audio generation model. - 根目录 > Video Generation > Wavespeed [openai/sora-2/image-to-video-pro](https://doc-en.302.ai/360390003e0.md): OpenAI's Sora 2 is a brand-new, state-of-the-art video and audio generation model. - 根目录 > Video Generation > Wavespeed [pixverse/pixverse-v5-t2v](https://doc-en.302.ai/361597674e0.md): AI finally becomes real in PixVerse V5. Smooth motion, natural details—all rendered in just seconds! - 根目录 > Video Generation > Wavespeed [pixverse/pixverse-v4.5-t2v](https://doc-en.302.ai/362162066e0.md): With the latest v4.5 model of PixVerse, you can generate high-quality videos based on text descriptions. This model supports multiple resolutions, aspect ratios, and motion modes, enabling flexible creation of various types of videos. - 根目录 > Video Generation > Wavespeed [pixverse/pixverse-v4.5-t2v-fast](https://doc-en.302.ai/362598485e0.md): With PixVerse's latest v4.5 model, you can generate high-quality videos based on text descriptions. It supports multiple resolutions, aspect ratios, and motion modes to meet diverse video creation needs. - 根目录 > Video Generation > Wavespeed [vidu/text-to-video-2.0](https://doc-en.302.ai/361512270e0.md): The Vidu text-to-video model can generate high-quality 720p videos with excellent visual effects and diverse dynamic performance. - 根目录 > Video Generation > Wavespeed [vidu/text-to-video-q1](https://doc-en.302.ai/362598481e0.md): The Vidu text-to-video generation model can produce high-quality videos with outstanding visual effects and a rich variety of dynamic expressions. - 根目录 > Video Generation > Wavespeed [wan-2.2-t2v-480p-ultra-fast](https://doc-en.302.ai/359725869e0.md): Generate infinite AI videos using the Wan 2.2 text-to-video model. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/speech-to-video](https://doc-en.302.ai/360390005e0.md): Wan-2.2-S2V is a video model that can generate high-quality videos based on static images and audio, presenting realistic facial expressions and body movements. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-720p-ultra-fast](https://doc-en.302.ai/360624799e0.md): By using the Wan 2.2 text-to-video generation model, you can create unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-720p](https://doc-en.302.ai/360624802e0.md): Use the Wan 2.2 text-to-video model to generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-5b-720p](https://doc-en.302.ai/360850246e0.md): ByteDance Seedance is an advanced AI video generation model that excels at producing coherent multi-shot videos. - 根目录 > Video Generation > Wavespeed [wan2.5 text-to-video-fast](https://doc-en.302.ai/359725865e0.md): Alibaba WAN 2.5 is a state-of-the-art text/image-to-video generation model available on Alibaba Cloud's DashScope platform. This model can generate high-quality videos with synchronized audio based on simple text or image prompts, supporting resolutions of 480p, 720p, or 1080p. Compared to Google Veo3, it offers higher cost-effectiveness and faster video generation speed, making it ideal for efficiently creating videos embedded with audio. This API provides a cost-effective solution for generating high-quality videos with synchronized audio from text prompts. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-480p-lora](https://doc-en.302.ai/360850247e0.md): Use the Wan 2.2 text-to-video model to generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/hunyuan-video/t2v](https://doc-en.302.ai/360850248e0.md): HunyuanVideo is an advanced text-to-video generation model that can create high-quality videos based on text descriptions. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-480p-lora-ultra-fast](https://doc-en.302.ai/361512268e0.md): Generate unlimited AI videos using the Wan 2.2 text-to-video model, with support for custom LoRA. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-720p-lora-ultra-fast](https://doc-en.302.ai/361512269e0.md): Using the Wan 2.2 text-to-video model, you can generate unlimited AI videos and customize LoRA. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-720p](https://doc-en.302.ai/361597671e0.md): With the Wan 2.1 text-to-video model, you can generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-480p](https://doc-en.302.ai/361597672e0.md): With the Wan 2.1 text-to-video model, you can generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/t2v-5b-720p-lora](https://doc-en.302.ai/361597673e0.md): With the Wan 2.2 text-to-video model, you can generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-720p-ultra-fast](https://doc-en.302.ai/362162067e0.md): Use the Wan 2.1 text-to-video model to generate unlimited AI videos. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-480p-lora](https://doc-en.302.ai/362162070e0.md): Using the Wan 2.1 text-to-video generation model, you can create unlimited AI videos and support custom LoRA (Low-Rank Adaptation) models for more flexible and personalized video generation effects. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-720p-lora](https://doc-en.302.ai/363672181e0.md): Generate unlimited AI videos using the Wan 2.1 text-to-video model, with support for custom LoRA models. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.1/t2v-480p-ultra-fast](https://doc-en.302.ai/362162071e0.md): Using the Wan 2.1 text-to-video generation model, you can create unlimited AI videos and customize LoRA (Low-Rank Adaptation) models for more flexible and personalized video generation effects. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/i2v-480p](https://doc-en.302.ai/363568413e0.md): Generate unlimited AI videos using the Wan 2.2 A14B image-to-video model. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2/i2v-720p](https://doc-en.302.ai/363568414e0.md): Generate unlimited AI videos using the Wan 2.2 A14B image-to-video model. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/infinitetalk](https://doc-en.302.ai/363568416e0.md): InfiniteTalk is an audio-driven conversational AI video generation model. With a single image and audio input, you can create dialogue or singing videos. Our pricing is $0.15 per 5 seconds of video generation (480p) or $0.3 (720p), and we support video generation up to 10 minutes in length. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/infinitetalk-fast](https://doc-en.302.ai/383587216e0.md): InfiniteTalk-Fast quickly converts a photo plus audio into an audio-driven talking or singing avatar video (image-to-video), producing videos with precise lip-sync and aligning head, facial, and body movements with the audio. It maintains identity consistency in infinitely long videos while supporting image-to-video generation, transforming static photos into vivid talking or singing videos lasting up to 10 minutes. A ready-to-use REST API is available, with no cold starts and affordable pricing. - 根目录 > Video Generation > Wavespeed [midjourney/image-to-video](https://doc-en.302.ai/368579431e0.md): Create stunning high-quality videos from images using Midjourney's powerful AI. Renowned for exceptional artistic quality and creative interpretation. - 根目录 > Video Generation > Wavespeed [lightricks/ltx-2-fast/image-to-video](https://doc-en.302.ai/368579432e0.md): LTX-2 is a complete AI creative engine. It is built for real production workflows and provides synchronized audio and video generation. - 根目录 > Video Generation > Wavespeed [lightricks/ltx-2-pro/image-to-video](https://doc-en.302.ai/368579433e0.md): LTX-2 is a comprehensive AI creative engine. Designed for real production workflows, it enables synchronized audio and video generation and supports 1080p. - 根目录 > Video Generation > Wavespeed [kwaivgi/kling-v2.5-turbo-std/image-to-video](https://doc-en.302.ai/368579434e0.md): kling-v2.5-turbo-std: Top-tier image-to-video generation with unparalleled motion smoothness, cinematic visual effects, and outstanding prompt accuracy. Now priced 25% lower compared to the 2.1 standard version, delivering exceptional quality at unbeatable value. - 根目录 > Video Generation > Wavespeed [vidu/reference-to-video-q2](https://doc-en.302.ai/368579435e0.md): Vidu Q2 is a new image-to-video (as well as reference-to-video) model that emphasizes subtle facial expressions and smooth dolly zoom movements. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2-spicy/image-to-video-lora](https://doc-en.302.ai/368579436e0.md): Wan 2.2 Spicy is a perfect video generation model for creating infinite content, featuring extremely high and smooth animation quality. This endpoint also supports loading custom LoRA weights. - 根目录 > Video Generation > Wavespeed [wavespeed-ai/wan-2.2-spicy/image-to-video](https://doc-en.302.ai/368579437e0.md): Wan 2.2 Spicy is a perfect video generation model that can produce unlimited content with extremely high quality and smooth animation. - 根目录 > Video Generation > Wavespeed [higgsfield/dop/image-to-video](https://doc-en.302.ai/368708776e0.md): The DoP service utilizes advanced AI motion synthesis technology to transform static images into dynamic 5-second videos. DoP takes the image you provide and applies realistic motion effects to create engaging video content. - 根目录 > Video Generation > Wavespeed [Wavespeed Retrieval Task](https://doc-en.302.ai/360631451e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Video Generation > Wavespeed [vidu/start-end-to-video-q2-turbo](https://doc-en.302.ai/371913593e0.md): Vidu Q2 Turbo Start-End to Video可在指定的起始和结束图像之间生成平滑过渡的视频。新模型效果好，生成速度快。 - 根目录 > Video Generation > Gongji Computing > Video Face Swapper [Create video face swap task](https://doc-en.302.ai/342848998e0.md): Create a video face-swapping task (replace the original face with the target face) - 根目录 > Video Generation > Gongji Computing > Video Face Swapper [View video face swap task](https://doc-en.302.ai/342848999e0.md): View Video Face Swap Task - 根目录 > Video Generation > Gaga [Video generation](https://doc-en.302.ai/383092872e0.md): This API can create tasks based on requirements and parameters, and return a task ID. - 根目录 > Video Generation > Gaga [Query](https://doc-en.302.ai/383093448e0.md): This interface can obtain results based on the ID of the video generation task - 根目录 > Video Generation > Gaga [Upload](https://doc-en.302.ai/383093710e0.md): An interface for uploading material files, where you can upload files you want to use as materials and obtain the corresponding identification ID. - 根目录 > Video Generation > Gaga [Get uploaded resources](https://doc-en.302.ai/383094393e0.md): This interface can query the uploaded files based on the provided identification id to ensure the traceability of the files. - 根目录 > Audio/Video Processing > Unified interface > TTS > 302 Format V1 [Text-to-Speech Generation (302 Format)](https://doc-en.302.ai/294943678e0.md): **Pricing: Charged according to the corresponding TTS provider** - 根目录 > Audio/Video Processing > Unified interface > TTS > 302 Format V2 [Text-to-Speech Generation (302 Format)](https://doc-en.302.ai/343760428e0.md): **Pricing: Charged according to the corresponding TTS provider** - 根目录 > Audio/Video Processing > Unified interface > TTS > 302 Format V2 [Webhook Request Example on Success](https://doc-en.302.ai/343760430e0.md): A 200 status response indicates that the webhook push data was successful. If it fails, it will retry three times with a delay. - 根目录 > Audio/Video Processing > Unified interface > TTS > 302 Format V2 [Query TTS Task](https://doc-en.302.ai/343760429e0.md): - 根目录 > Audio/Video Processing > Unified interface > TTS > Openai Format [Text-to-Speech Generation (Openai Format)](https://doc-en.302.ai/328166279e0.md): Model/Voice List & Parameter Description: - 根目录 > Audio/Video Processing > Unified interface > TTS [Query TTS Provider Info](https://doc-en.302.ai/294943728e0.md): - 根目录 > Audio/Video Processing > 302.AI > IndexTTS-2 [Create TTS Task](https://doc-en.302.ai/350161039e0.md): **Price: 0.015PTC/1k tokens, minimum charge 0.001PTC** - 根目录 > Audio/Video Processing > 302.AI > IndexTTS-2 [Query Task](https://doc-en.302.ai/350161040e0.md): - 根目录 > Audio/Video Processing > 302.AI > Higgs Audio [Create Voice Cloning Task](https://doc-en.302.ai/338003333e0.md): Create Voice Cloning - 根目录 > Audio/Video Processing > 302.AI > Higgs Audio [View Voice Cloning Tasks](https://doc-en.302.ai/338003334e0.md): View Voice Cloning Tasks - 根目录 > Audio/Video Processing > 302.AI > Higgs Audio [Create Smart Voice Generation Task](https://doc-en.302.ai/338003335e0.md): Create an intelligent voice generation task, where the tone can be described using system_prompt - 根目录 > Audio/Video Processing > 302.AI > Higgs Audio [View Smart Voice Generation Tasks](https://doc-en.302.ai/338003336e0.md): View Intelligent Voice Generation Tasks - 根目录 > Audio/Video Processing > 302.AI > F5-TTS [F5-TTS（Text to Speech）](https://doc-en.302.ai/225254060e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Audio/Video Processing > 302.AI > F5-TTS [F5-TTS （Asynchronous Text-to-Speech）](https://doc-en.302.ai/244081300e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Audio/Video Processing > 302.AI > F5-TTS [F5-TTS （Asynchronously Retrieve Results）](https://doc-en.302.ai/244081628e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Audio/Video Processing > 302.AI > MMAudio [mmaudio（Text-to-Speech）](https://doc-en.302.ai/248597785e0.md): AI video dubbing with text input, voice generation, and synchronized API - 根目录 > Audio/Video Processing > 302.AI > MMAudio [mmaudio（AI Video Voiceover）](https://doc-en.302.ai/245622362e0.md): AI Video Voiceover - 根目录 > Audio/Video Processing > 302.AI > MMAudio [mmaudio （Asynchronous Result Retrieval）](https://doc-en.302.ai/245622558e0.md): AI Video Voiceover - 根目录 > Audio/Video Processing > 302.AI > SoulX-Podcast [Create TTS task](https://doc-en.302.ai/373210516e0.md): A TTS model from SoulX, supporting quick voice cloning and generating multi-person voice conversations - 根目录 > Audio/Video Processing > 302.AI > SoulX-Podcast [Create Podcast task](https://doc-en.302.ai/373210517e0.md): The TTS model from SoulX supports quick voice cloning and generating multi-person voice conversations. - 根目录 > Audio/Video Processing > 302.AI > SoulX-Podcast [Query task](https://doc-en.302.ai/373210518e0.md): - 根目录 > Audio/Video Processing > 302.AI > VoxCPM-TTS [Create TTS task](https://doc-en.302.ai/360840697e0.md): **Price: 0.2PTC/1k tokens** - 根目录 > Audio/Video Processing > 302.AI > VoxCPM-TTS [Query task](https://doc-en.302.ai/360840698e0.md): - 根目录 > Audio/Video Processing > 302.AI > Audio to Text [Transcriptions（Speech to Text）](https://doc-en.302.ai/207705229e0.md): Transcribe audio into the input language. - 根目录 > Audio/Video Processing > 302.AI > Audio to Text [Transcript (Audio/Video to Text)](https://doc-en.302.ai/207705230e0.md): Automatically extract speech from video or audio and convert it into text subtitles. - 根目录 > Audio/Video Processing > 302.AI > Audio to Text [Alignments（Subtitle Timing） ](https://doc-en.302.ai/243529129e0.md): - 根目录 > Audio/Video Processing > 302.AI > Audio to Text [WhisperX](https://doc-en.302.ai/238598485e0.md): "Open-source version"WhsiperX - 根目录 > Audio/Video Processing > 302.AI > Video related [Video-Utile（Get task results）](https://doc-en.302.ai/362135065e0.md): 302.AI's API comes from models we have deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by us. - 根目录 > Audio/Video Processing > 302.AI > Video related [Video-Utils](https://doc-en.302.ai/362135064e0.md): Video-Utile is a powerful video processing tool API that offers a variety of video editing and conversion features. This API uses a pay-per-use time-based billing model, providing users with flexible and efficient video processing services. - 根目录 > Audio/Video Processing > 302.AI > Video related [Video-Understanding (Video understanding)](https://doc-en.302.ai/331356566e0.md): Input video, output video understanding, asynchronous API - 根目录 > Audio/Video Processing > 302.AI > Video related [Video-Understanding (Asynchronously get result)](https://doc-en.302.ai/331356567e0.md): AI Video Understanding - 根目录 > Audio/Video Processing > 302.AI > Audio translation [Create an audio translation task](https://doc-en.302.ai/379775312e0.md): # Function Introduction - 根目录 > Audio/Video Processing > 302.AI > Audio translation [Query audio translation task](https://doc-en.302.ai/379779193e0.md): - 根目录 > Audio/Video Processing > 302.AI > Others [Stable-Audio（instrumental generation）](https://doc-en.302.ai/219194446e0.md): 302.AI's API comes from models we deployed on cloud GPUs. Some models are open-source, while others are fine-tuned or developed by 302.AI. - 根目录 > Audio/Video Processing > 302.AI > Others [Diffrhythm（Song Generation）](https://doc-en.302.ai/268457593e0.md): Open-source song generation requires inputting reference music and lyrics to generate music. The maximum length generated is 1 minute and 35 seconds. Currently, the quality of English songs is better than that of Chinese songs. - 根目录 > Audio/Video Processing > OpenAI [Speech（Text to Speech tts-1）](https://doc-en.302.ai/207705220e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Audio/Video Processing > OpenAI [Transcriptions（Speech to Text）](https://doc-en.302.ai/207705218e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Audio/Video Processing > OpenAI [Translations（Speech to English Text whisper-1）](https://doc-en.302.ai/207705219e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > Audio/Video Processing > OpenAI [Realtime](https://doc-en.302.ai/222610017e0.md): [Official Guide](https://platform.openai.com/docs/guides/realtime/quickstart) - 根目录 > Audio/Video Processing > Azure [AzureTTS（Text to Speech）](https://doc-en.302.ai/207705221e0.md): **Text-to-Speech service provided by Microsoft Azure** - 根目录 > Audio/Video Processing > Azure [Voice-List](https://doc-en.302.ai/207705222e0.md): **Text-to-Speech service provided by Microsoft Azure** - 根目录 > Audio/Video Processing > Suno [Music（Automatic Mode）](https://doc-en.302.ai/207705226e0.md): **Enter a keyword, automatically generate a song** - 根目录 > Audio/Video Processing > Suno [Music（Custom Mode）](https://doc-en.302.ai/207705223e0.md): Generate 2 songs by customizing settings such as lyrics and style of the songs. - 根目录 > Audio/Video Processing > Suno [Music（Song Continuation）](https://doc-en.302.ai/207705224e0.md): Based on the previously generated song, continue writing a new song. You can set the continuation duration, with each continuation fixed at 2 minutes. - 根目录 > Audio/Video Processing > Suno [Music（Generate Lyrics）](https://doc-en.302.ai/207705227e0.md): **Enter a keyword, automatically generate lyrics.** - 根目录 > Audio/Video Processing > Suno [Upload (Upload Material)](https://doc-en.302.ai/370442529e0.md): Upload materials for later use - 根目录 > Audio/Video Processing > Suno [Add Instrumental (Add Instrumental to A Cappella)](https://doc-en.302.ai/370442530e0.md): Add Instrumental feature released by suno in August 2025 - 根目录 > Audio/Video Processing > Suno [Add Vocals (Add Vocals to Instrumental)](https://doc-en.302.ai/370442531e0.md): Add Vocals Feature Released by suno in August 2025 - 根目录 > Audio/Video Processing > Suno [Timing (Lyrics Timeline)](https://doc-en.302.ai/370442532e0.md): Obtain the lyrics timeline, for example: - 根目录 > Audio/Video Processing > Suno [Wav (Get Wav File)](https://doc-en.302.ai/370442533e0.md): Get Wav File - 根目录 > Audio/Video Processing > Suno [Video (Get Video)](https://doc-en.302.ai/370442534e0.md): Get Video - 根目录 > Audio/Video Processing > Suno [Speed (Adjust Song Speed)](https://doc-en.302.ai/370442535e0.md): Adjust Song Speed - 根目录 > Audio/Video Processing > Suno [Crop (Trim Music)](https://doc-en.302.ai/370442536e0.md): Crop Music - 根目录 > Audio/Video Processing > Suno [Fetch](https://doc-en.302.ai/207705225e0.md): Query the song generation status. - 根目录 > Audio/Video Processing > Doubao [tts_hd（Text to Speech）](https://doc-en.302.ai/207705231e0.md): **Text-to-Speech API from Doubao** - 根目录 > Audio/Video Processing > Doubao [vc-ata（Automatic subtitle timing）](https://doc-en.302.ai/241439484e0.md): Automatic subtitle timing from Doubao. - 根目录 > Audio/Video Processing > Doubao [fetch（Query Generation Status）](https://doc-en.302.ai/241438924e0.md): Automatic subtitle timing from Doubao，query generation status - 根目录 > Audio/Video Processing > Doubao [vc（Audio and video caption generation）](https://doc-en.302.ai/242584438e0.md): Audio and video caption generation from Doubao - 根目录 > Audio/Video Processing > Doubao [fetch（Query caption result）](https://doc-en.302.ai/242593526e0.md): Automatic caption timing from Doubao, check generation status - 根目录 > Audio/Video Processing > Doubao [Recognize (Rapid Audio File Recognition)](https://doc-en.302.ai/330758453e0.md): This API is suitable for rapid recognition scenarios of audio recordings. Based on large model capabilities, it provides better recognition results and faster response times. The invocation method is a single request that returns the recognition result immediately, without the need for submit/query polling. - 根目录 > Audio/Video Processing > Fish Audio [TTS（Text to Speech）](https://doc-en.302.ai/216956018e0.md): Text-to-speech from Fish Audio - 根目录 > Audio/Video Processing > Fish Audio [Model（Create Voice）](https://doc-en.302.ai/216956017e0.md): Sound cloning from Fish Audio: submit audio files for cloning. - 根目录 > Audio/Video Processing > Fish Audio [Model（Obtain Voice）](https://doc-en.302.ai/216956019e0.md): Sound cloning from Fish Audio - 根目录 > Audio/Video Processing > Fish Audio [Model（Delete Voice）](https://doc-en.302.ai/216956021e0.md): Sound cloning from Fish Audio - 根目录 > Audio/Video Processing > Fish Audio [Model（Update Voice）](https://doc-en.302.ai/216956022e0.md): Sound cloning from Fish Audio - 根目录 > Audio/Video Processing > Fish Audio [Model（Get Voice List）](https://doc-en.302.ai/216956020e0.md): Public Sound List from Fish Audio - 根目录 > Audio/Video Processing > Minimax [T2A（Speech Generation - Synchronous）](https://doc-en.302.ai/336008006e0.md): Text-to-Speech Tone Frequency from Minimax - 根目录 > Audio/Video Processing > Minimax [T2A（Async extra content generation）](https://doc-en.302.ai/224542216e0.md): Text-to-Audio Tone Frequency from Minimax - 根目录 > Audio/Video Processing > Minimax [T2A（Status Inquiry）](https://doc-en.302.ai/225432825e0.md): Vincennes Tone Frequency from Minimax， - 根目录 > Audio/Video Processing > Minimax [T2V（Create Voice）](https://doc-en.302.ai/224542215e0.md): Vincennes Tone Frequency from Minimax - 根目录 > Audio/Video Processing > Minimax [Files（Audio File Download）](https://doc-en.302.ai/224542217e0.md): Vincennes Video Model from Minimax - 根目录 > Audio/Video Processing > Minimax [Music Generation API](https://doc-en.302.ai/241838654e0.md): Allow users to generate high-quality AI music with natural language inspiration descriptions (prompt) and complete lyrics (lyrics) with one click. The currently supported model is **music-1.5** **music-2.0**, which is suitable for music composition, lyric matching, AI-generated vocals, and other scenarios. - 根目录 > Audio/Video Processing > Minimax [Upload](https://doc-en.302.ai/328237587e0.md): If you need to clone a voice, please upload it through this endpoint first. - 根目录 > Audio/Video Processing > Minimax [Voice Clone](https://doc-en.302.ai/328166280e0.md): Clone audition parameters. The model will use the cloned voice to read the provided text content, and return the synthesized audio result as a link for auditioning the cloning effect. Limited to 2000 characters. Note: Audition will incur normal speech synthesis charges based on character count, with pricing consistent with all T2A interfaces. - 根目录 > Audio/Video Processing > Dubbingx [TTS（Text to Speech）](https://doc-en.302.ai/225432829e0.md): Text-to-speech asynchronous interface - 根目录 > Audio/Video Processing > Dubbingx [GetTTSList（Get Voice List）](https://doc-en.302.ai/225432826e0.md): Getting the Tone List - 根目录 > Audio/Video Processing > Dubbingx [GetTTSTask（Get Task Status）](https://doc-en.302.ai/225432827e0.md): Getting Task Status - 根目录 > Audio/Video Processing > Dubbingx [Analyze（emotions）](https://doc-en.302.ai/225432828e0.md): Optionally, analyze sentiment based on the text and return the results. - 根目录 > Audio/Video Processing > Udio [Generate（Music Generation）](https://doc-en.302.ai/231604784e0.md): Enter prompt and generate a song - 根目录 > Audio/Video Processing > Udio [Generate（Music Continuation）](https://doc-en.302.ai/232137740e0.md): A continuation of the song - 根目录 > Audio/Video Processing > Udio [Query](https://doc-en.302.ai/231604783e0.md): Checking Song Generation - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [Speech-to-text（Speech-to-Text）](https://doc-en.302.ai/270141046e0.md): Speech-to-text from ElevenLabs, featuring the ability to mark applause, laughter, etc. - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [Speech-to-text（Asynchronously fetch results）](https://doc-en.302.ai/270143269e0.md): Elevenlabs' speech-to-text feature can mark applause, laughter, and more. - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Multilingual-v2（文字转语音同步）](https://doc-en.302.ai/338307765e0.md): Elevenlabsのテキスト音声変換 - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Multilingual-v2（Text-to-Speech）](https://doc-en.302.ai/279192639e0.md): From Elevenlabs Text-to-Speech - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Multilingual-v2（Asynchronous result retrieval）](https://doc-en.302.ai/279192970e0.md): From Elevenlabs Text-to-Speech - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Flash-v2.5（文字转语音同步）](https://doc-en.302.ai/338307766e0.md): Elevenlabsのテキスト音声変換 - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Flash-v2.5（Text-to-Speech）](https://doc-en.302.ai/279193033e0.md): From Elevenlabs Text-to-Speech - 根目录 > Audio/Video Processing > Elevenlabs > 302 Format [TTS-Flash-v2.5（Asynchronous result retrieval）](https://doc-en.302.ai/279193050e0.md): From Elevenlabs Text-to-Speech - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Text-to-speech](https://doc-en.302.ai/316945684e0.md): From Elevenlabs Text-to-Speech - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Speech-to-text](https://doc-en.302.ai/318268277e0.md): Speech-to-text from Elevenlabs, we have modified the API and also support direct reading from URL addresses. - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Text-to-Dialogue (Create Multi-person Dialogue)](https://doc-en.302.ai/340677701e0.md): Use the eleven_v3 model to convert a list of text and voice ID pairs into a multi-speaker dialogue and return the audio. - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Music (Music Generation)](https://doc-en.302.ai/340677702e0.md): Compose a song based on a prompt or composition plan. - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Models (Get Models)](https://doc-en.302.ai/316945685e0.md): Get the model ID corresponding to text-to-speech - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Voices（Get Voices）](https://doc-en.302.ai/316945686e0.md): Get the voice_id corresponding to text-to-speech - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Audio-isolation (audio isolation)](https://doc-en.302.ai/385578428e0.md): Audio isolation from Elevenlabs, removing background noise from audio. - 根目录 > Audio/Video Processing > Elevenlabs > Official Format [Sound-generation (sound effect generation)](https://doc-en.302.ai/385578540e0.md): Sound effects from Elevenlabs, which convert text into sound effects for videos, dubbing, or video games, using the world's most advanced sound effect models. - 根目录 > Audio/Video Processing > Mureka [Upload Music](https://doc-en.302.ai/281661555e0.md): Upload a file that can be used across multiple different endpoints. - 根目录 > Audio/Video Processing > Mureka [Generate Lyrics from a Prompt](https://doc-en.302.ai/281661796e0.md): Create lyrics based on a given prompt. - 根目录 > Audio/Video Processing > Mureka [Continue writing lyrics from existing lyrics](https://doc-en.302.ai/281661968e0.md): Keep extending the lyrics from the current ones. - 根目录 > Audio/Video Processing > Mureka [Generate a Song from Lyrics](https://doc-en.302.ai/281661992e0.md): Generate songs based on lyrics. - 根目录 > Audio/Video Processing > Mureka [Retrieve the Generated Song](https://doc-en.302.ai/281662029e0.md): Fetch the generated song. - 根目录 > Audio/Video Processing > Mureka [Separate Music Stems](https://doc-en.302.ai/281662109e0.md): Separate the input music into individual audio elements such as vocals, instruments, etc. - 根目录 > Audio/Video Processing > Mureka [Generate Instrumental Music Track](https://doc-en.302.ai/281662147e0.md): Generate instrumental music based on user input. - 根目录 > Audio/Video Processing > Mureka [Retrieve Instrumental Music Track](https://doc-en.302.ai/281662169e0.md): Generate instrumental music based on user input. - 根目录 > Audio/Video Processing > Mureka [Text-to-Speech](https://doc-en.302.ai/281662174e0.md): Generate audio from input text. - 根目录 > Audio/Video Processing > Mureka [Create Podcast Audio](https://doc-en.302.ai/281662201e0.md): Convert a two-person dialogue script into a natural-sounding, podcast-style audio conversation ready for publishing. - 根目录 > Audio/Video Processing > SiliconFlow [Upload reference audio](https://doc-en.302.ai/308965990e0.md): Supported models: FunAudioLLM/CosyVoice2-0.5B - 根目录 > Audio/Video Processing > SiliconFlow [Delete reference audio](https://doc-en.302.ai/308965991e0.md): Delete reference audio, value taken from the audio list - 根目录 > Audio/Video Processing > SiliconFlow [Create speech-to-text request](https://doc-en.302.ai/308965992e0.md): Model supported: FunAudioLLM/SenseVoiceSmall - 根目录 > Audio/Video Processing > SiliconFlow [Create text-to-speech request](https://doc-en.302.ai/308965993e0.md): Supported model: FunAudioLLM/CosyVoice2-0.5B - 根目录 > Audio/Video Processing > SiliconFlow [FunAudioLLM/CosyVoice2-0.5B TTS](https://doc-en.302.ai/333523947e0.md): CosyVoice 2 is a streaming speech synthesis model based on large language models, featuring a unified framework design for both streaming and non-streaming modes. The model enhances the encoding efficiency of speech tokens through Finite Scalar Quantization (FSQ), simplifies the language model architecture for text-to-speech, and introduces a chunk-aware causal flow matching model to support various synthesis scenarios. In streaming mode, the model achieves ultra-low latency of 150ms while maintaining nearly the same synthesis quality as the non-streaming mode. - 根目录 > Audio/Video Processing > SiliconFlow [fnlp/MOSS-TTSD-v0.5 TTS](https://doc-en.302.ai/333523948e0.md): MOSS-TTSD (text to spoken dialogue) is an open-source bilingual (supports Chinese and English) spoken dialogue synthesis model. It can convert dialogue scripts between two speakers into natural and expressive conversational speech. Its main highlights include: generating highly expressive, human-like dialogue speech; supporting zero-shot two-speaker voice cloning; and generating long-form speech up to 960 seconds in a single pass. The model is built upon a unified semantic-acoustic neural audio codec, a pre-trained large language model, and massive speech data, making it ideal for scenarios such as AI podcast production. - 根目录 > Audio/Video Processing > Google [Text-to-Speech](https://doc-en.302.ai/310167102e0.md): From Google's TTS model, reference: https://cloud.google.com/text-to-speech#demo - 根目录 > Audio/Video Processing > Google [gemini-2.5-flash-preview-tts](https://doc-en.302.ai/310167103e0.md): Supported model: gemini-2.5-flash-preview-tts - 根目录 > Audio/Video Processing > Google [gemini-2.5-pro-preview-tts](https://doc-en.302.ai/310167104e0.md): Supported model: gemini-2.5-pro-preview-tts - 根目录 > Audio/Video Processing > Chanjing [Create a language generation task](https://doc-en.302.ai/312102531e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/QOG9lyrgJP3Qp73mHnE1MQ4YVzN67Mw4 - 根目录 > Audio/Video Processing > Chanjing [Retrieve speech synthesis results](https://doc-en.302.ai/312102532e0.md): | Primary Field | Secondary Field | Tertiary Field | Description | - 根目录 > Audio/Video Processing > Chanjing [Create a voice customization task](https://doc-en.302.ai/312102533e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/AR4GpnMqJzM6gPMwuoalO5ZRVKe0xjE3 - 根目录 > Audio/Video Processing > Chanjing [Retrieve voice customization results](https://doc-en.302.ai/312102534e0.md): Reference: https://alidocs.dingtalk.com/i/nodes/ydxXB52LJq7yOZ7zuEMY2vYzWqjMp697 - 根目录 > Audio/Video Processing > Chanjing [Delete customized voice](https://doc-en.302.ai/312102535e0.md): - 根目录 > Audio/Video Processing > Kling [Video-to-Audio (Video Sound Effects Generation)](https://doc-en.302.ai/340677703e0.md): **Price: 0.05 PTC/time** - 根目录 > Audio/Video Processing > Kling [Video-to-Audio (Get Task Results)](https://doc-en.302.ai/340677704e0.md): This interface is only used to obtain tasks and is free of charge. - 根目录 > Audio/Video Processing > Kling [Text-to-Audio (Text to Sound Effects)](https://doc-en.302.ai/340677705e0.md): **Price: 0.05 PTC/time** - 根目录 > Audio/Video Processing > Kling [Text-to-Audio (Get Task Results)](https://doc-en.302.ai/340677706e0.md): This interface is only used to obtain tasks and is free of charge. - 根目录 > Audio/Video Processing > Tongyi Wanxiang [Qwen-TTS (Speech Synthesis)](https://doc-en.302.ai/342267513e0.md): Qwen-TTS is a speech synthesis model from the Tongyi Qianwen series. It supports input of Chinese, English, and mixed Chinese-English text, and outputs audio in a streaming manner. - 根目录 > Audio/Video Processing > Tongyi Wanxiang [Qwen3-TTS-Flash（Speech Synthesis）](https://doc-en.302.ai/354933210e0.md): Qwen3-TTS-Flash is a speech synthesis model from the Qwen series. Qwen3-TTS offers 17 voice styles and supports multiple languages and dialects. - 根目录 > Audio/Video Processing > Topazlabs [Video enhancement to high definition](https://doc-en.302.ai/356457906e0.md): Video enhancement by Topazlabs: Use AI to upscale blurry videos to 4K. - 根目录 > Audio/Video Processing > Topazlabs [Retrieve task result](https://doc-en.302.ai/356457907e0.md): **Price: Free** - 根目录 > Audio/Video Processing > Stability [Text-to-Audio (Text-generated Music)](https://doc-en.302.ai/357015914e0.md): Stable Audio can generate up to three minutes of high-quality music and sound effects based on text descriptions, with a 44.1kHz stereo sampling rate. Refer to our prompt guide to learn how to write effective prompts for the best generation results. - 根目录 > Audio/Video Processing > Stability [Audio-to-Audio (Reference-based Music Generation)](https://doc-en.302.ai/357015915e0.md): Stable Audio can transform existing audio samples into new high-quality works, up to three minutes long, in 44.1kHz stereo, using text prompts. Explore sample transformation techniques in our "Audio-to-Audio Guide" to maximize creative control. Read more about model capabilities here. - 根目录 > Audio/Video Processing > Stability [Inpaint (Music Editing/Modification)](https://doc-en.302.ai/357015916e0.md): Stable Audio 2.5 uses text prompts to transform existing audio samples into high-quality works up to three minutes long, outputting in 44.1kHz stereo. - 根目录 > Audio/Video Processing > Wavespeed > General Interface [Wavespeed Generation Task](https://doc-en.302.ai/360313611e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 at the same price. - 根目录 > Audio/Video Processing > Wavespeed > General Interface [Wavespeed Retrieval Task](https://doc-en.302.ai/360313612e0.md): 302 and Wavespeed have officially partnered, and now all Wavespeed models are available on 302 with no change in price. - 根目录 > Audio/Video Processing > Wavespeed [wavespeed-ai/infinitetalk-fast/video-to-video](https://doc-en.302.ai/383620708e0.md): infiniteTalk/video-to-vedio creates new videos by combining input silent videos and audio. It ensures precise lip-sync while aligning head, facial, and body movements with the audio. With optional masks and prompts, users can control which areas move and how the scene is presented. The model also maintains visual recognition for natural and consistent results. A ready-to-use REST inference API delivers optimal performance, with no cold starts and affordable pricing. - 根目录 > Audio/Video Processing > Zhipu [GLM-TTS](https://doc-en.302.ai/391676977e0.md): The GLM-TTS speech synthesis model, with the new-generation Zhipu speech large model at its core, breaks through the traditional speech synthesis framework. It intelligently predicts the emotion and intonation of the text through context, significantly improving the naturalness and expressiveness of the speech, making the synthesized speech possess real emotions and vitality. GLM-TTS adopts a two-stage generation in its architecture and introduces a reinforcement learning scheme based on GRPO during training, achieving open-source SOTA performance in the publicly evaluated "character error rate" and "emotional expression". - 根目录 > Audio/Video Processing > Zhipu [GLM-TTS-Clone](https://doc-en.302.ai/391677199e0.md): GLM-TTS-Clone is a voice cloning model launched by Zhipu. With just a 3-second voice sample, it can learn the speaker's timbre and speech habits, and can generate natural, smooth, and human-like voices in scenarios such as general reading, emotional dubbing, educational evaluation, e-books, and voice customer service. - 根目录 > Audio/Video Processing > Zhipu [GLM-ASR-2512](https://doc-en.302.ai/391333046e0.md): GLM-ASR-2512 is Zhipu's new-generation speech recognition model, which supports the real-time conversion of speech into high-quality text. Whether it is daily chat, meeting minutes, work documents, or scenarios involving professional terms, it can accurately recognize and convert, greatly improving the efficiency of input and recording. The model maintains industry-leading recognition performance under multi-scenario and multi-accent conditions, with a Character Error Rate (CER) of only 0.0717, providing a fast and reliable speech input experience. - 根目录 > Audio/Video Processing > Zhipu [Upload files](https://doc-en.302.ai/391677370e0.md): Upload files for functions such as `Batch` and `Agent`. Note that the `Try it` function only supports uploading small files. For the actual supported file size, please refer to the relevant description of `purpose` below. - 根目录 > Information Processing > Unified Search API [Unified Search API](https://doc-en.302.ai/293665533e0.md): **Price: Charged according to the corresponding search provider** - 根目录 > Information Processing > 302.AI > Admin Dashboard [Balance（Account balance）](https://doc-en.302.ai/263735171e0.md): Get the balance of the corresponding 302.AI account - 根目录 > Information Processing > 302.AI > Admin Dashboard [Record（Billing Details）](https://doc-en.302.ai/358337598e0.md): Query the deduction for each request based on the request-id in the request header. - 根目录 > Information Processing > 302.AI > Admin Dashboard [API-Record（API Log Query）](https://doc-en.302.ai/388890877e0.md): Query recent request logs and deduction status corresponding to the API-KEY - 根目录 > Information Processing > 302.AI > Admin Dashboard [Price（Get API Pricing）](https://doc-en.302.ai/295927334e0.md): Obtain the model price for a specific path by specifying the Path parameter - 根目录 > Information Processing > 302.AI > Admin Dashboard [Retrieve User API Keys List Data](https://doc-en.302.ai/294542491e0.md): Retrieve the list of user API Keys using an API Key with system permissions. - 根目录 > Information Processing > 302.AI > Admin Dashboard [Retrieve Data for Specified API Key](https://doc-en.302.ai/294542600e0.md): Retrieve data for a specified API Key using an API Key with system permissions. - 根目录 > Information Processing > 302.AI > Admin Dashboard [Create API Key](https://doc-en.302.ai/294542670e0.md): Create an API Key using an API Key with system permissions. - 根目录 > Information Processing > 302.AI > Admin Dashboard [Update API Key](https://doc-en.302.ai/294542769e0.md): Update a specified API Key using an API Key with system permissions. - 根目录 > Information Processing > 302.AI > Admin Dashboard [Delete API Key](https://doc-en.302.ai/294543052e0.md): Delete a specified API Key name using an API Key with system permissions. - 根目录 > Information Processing > 302.AI > Admin Dashboard [API-SSO](https://doc-en.302.ai/404803089e0.md): **Price: Free** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Search](https://doc-en.302.ai/214738984e0.md): **Price：0.02PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Search (Xiaohongshu Notes Search V2)](https://doc-en.302.ai/341489259e0.md): **Price: 0.02PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Search (Xiaohongshu Notes Search V3)](https://doc-en.302.ai/341489260e0.md): **Price: 0.02PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Note (Xiaohongshu Note Retrieval)](https://doc-en.302.ai/341869378e0.md): **Price: 0.01PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Note (Xiaohongshu Note Retrieval V2)](https://doc-en.302.ai/214738985e0.md): **Price: 0.01PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Note (Xiaohongshu Note Retrieval V3)](https://doc-en.302.ai/341868679e0.md): **Price: 0.01PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Xiaohongshu_Comments](https://doc-en.302.ai/357304847e0.md): **Price: 0.01 PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Get_Home_Recommend](https://doc-en.302.ai/238638370e0.md): **Price：0.01PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Tiktok_Search](https://doc-en.302.ai/214738989e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Douyin_Search](https://doc-en.302.ai/214738990e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Twitter_Search](https://doc-en.302.ai/236434650e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Twitter_Post（X_Post）](https://doc-en.302.ai/214738986e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Twitter_User（X_User）](https://doc-en.302.ai/214738988e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Weibo_Post](https://doc-en.302.ai/214738987e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Search_Video](https://doc-en.302.ai/236434477e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Youtube_Info](https://doc-en.302.ai/222610812e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Youtube_Subtitles（Youtube Obtain Subtitles）](https://doc-en.302.ai/252702346e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [Bilibili_Info（Bilibili Obtain Video Information）](https://doc-en.302.ai/252702200e0.md): **Price：0.001PTC/call** - 根目录 > Information Processing > 302.AI > Information search [MP_Article_List（Get the list of WeChat official account articles）](https://doc-en.302.ai/269224841e0.md): **Price: 0.01 PTC/call** - 根目录 > Information Processing > 302.AI > Information search [MP_Article（Retrieve WeChat Official Account articles）](https://doc-en.302.ai/269233916e0.md): **价格：0.001PTC/次** - 根目录 > Information Processing > 302.AI > Information search [Zhihu_AI_Search (Zhihu AI Search)](https://doc-en.302.ai/342047190e0.md): **Price: 0.001PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Zhihu_AI_Search (Retrieve Zhihu AI Search Results)](https://doc-en.302.ai/342047191e0.md): **Price: 0.001PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Zhihu_Hot_List (Zhihu Hot List / Trending Topics)](https://doc-en.302.ai/342047192e0.md): **Price: 0.001PTC/time** - 根目录 > Information Processing > 302.AI > Information search [Video_Data (Retrieve Video Data)](https://doc-en.302.ai/342047193e0.md): **Price: 0.001PTC/time** - 根目录 > Information Processing > 302.AI > File processing [Parsing](https://doc-en.302.ai/226751725e0.md): Converting files to text format for streamlined processing of large-scale models - 根目录 > Information Processing > 302.AI > File processing [Upload-File](https://doc-en.302.ai/232502112e0.md): Upload the file for the LLM to process. - 根目录 > Information Processing > 302.AI > File processing [Markitdown (File conversion to md format)](https://doc-en.302.ai/307655879e0.md): Convert other files to md files - 根目录 > Information Processing > 302.AI > File processing [Link-to-IMG (Convert Link to Image)](https://doc-en.302.ai/360266558e0.md): **Price: 0.001 PTC/time** - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [One-click Code Execution](https://doc-en.302.ai/276825891e0.md): Automatically create a sandbox, and destroy it immediately after execution. Optional feature to export sandbox files (if there are multiple files in the directory, they will be compressed into a zip file for export; a single file will be exported directly). This interface is recommended if continuous sandbox operations are not required. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Create Sandbox](https://doc-en.302.ai/276825984e0.md): After successful creation, the sandbox will automatically pause. When you call other sandbox operation interfaces, the sandbox will automatically reconnect, and after execution, it will pause again to avoid unnecessary costs. (Note: Pausing and reconnecting will take some time, approximately 5 seconds in total.) - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Query Your Sandbox List](https://doc-en.302.ai/276826458e0.md): Sandbox information is bound to the API key, so you can only query the sandbox information associated with the current API key. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Destroy Sandbox](https://doc-en.302.ai/276826507e0.md): **Price：0 PTC/call** - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Run-Code](https://doc-en.302.ai/276828474e0.md): This interface only returns text-type outputs. If the code involves file generation or similar operations, please use the "View File" interface to check file information and the "Export File" interface to export files. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Run Command Line](https://doc-en.302.ai/276829298e0.md): This interface only returns text-type outputs. If the command involves file generation or similar operations, please use the "View File" interface to check file information and the "File Download" interface to export files to the 302 file system. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Query File Information at Specified Path](https://doc-en.302.ai/276829634e0.md): Supports batch queries; you can pass a list of paths. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Import File Data into Sandbox](https://doc-en.302.ai/276829674e0.md): Supports batch import. If a file exists at the save path, it will be overwritten. If the folder at the save path does not exist, it will be automatically created. - 根目录 > Information Processing > 302.AI > Code execution > Virtual Machine Sandbox [Export Sandbox Files](https://doc-en.302.ai/276830648e0.md): Supports Batch Export - 根目录 > Information Processing > 302.AI > Code execution > Static Sandbox [Run-Code](https://doc-en.302.ai/239070004e0.md): Run JS or Python code in a sandbox - 根目录 > Information Processing > 302.AI > Remote Browser [Asynchronously create browser automation tasks](https://doc-en.302.ai/342541289e0.md): Create an automation task for a remote browser based on Browser Use - 根目录 > Information Processing > 302.AI > Remote Browser [Create Browser Automation Task](https://doc-en.302.ai/282235063e0.md): Create a remote browser automation task based on Browser Use. - 根目录 > Information Processing > 302.AI > Remote Browser [Query Browser Task Status](https://doc-en.302.ai/282235713e0.md): Create a remote browser automation task based on Browser Use. - 根目录 > Information Processing > 302.AI > Paper2Code [Create paper2poster task (pass parameters via JSON)](https://doc-en.302.ai/342659153e0.md): ### Create a Paper2Poster Task - 根目录 > Information Processing > 302.AI > Paper2Code [Create Paper2Code Task](https://doc-en.302.ai/315837727e0.md): Create a Paper2Code task, based on the paper: https://arxiv.org/abs/2504.17192 - 根目录 > Information Processing > 302.AI > Paper2Code [Query Paper2Code Task](https://doc-en.302.ai/315837728e0.md): Query Paper2Code task - 根目录 > Information Processing > 302.AI > Paper2Poster [Create a paper2code task (pass parameters in JSON)](https://doc-en.302.ai/342583494e0.md): Create a paper2code task from the paper: https://arxiv.org/abs/2504.17192 - 根目录 > Information Processing > 302.AI > Paper2Poster [Create paper2poster Task](https://doc-en.302.ai/318268278e0.md): Create a paper2poster task, from the open-source project: https://paper2poster.github.io/ - 根目录 > Information Processing > 302.AI > Paper2Poster [Query paper2poster Task](https://doc-en.302.ai/318268279e0.md): Query the paper2poster task. The final output is a PPT file with both images and text, which is convenient for editing. - 根目录 > Information Processing > 302.AI > LLMxMapReduce [Create Writing Task](https://doc-en.302.ai/327572455e0.md): Create llm_mr_v2 Writing Task - 根目录 > Information Processing > 302.AI > LLMxMapReduce [Query Writing Task](https://doc-en.302.ai/327572456e0.md): Querying the llm_mr_v2 writing task takes about 15 minutes per task, which is relatively long. After completion, a Markdown file will be returned. - 根目录 > Information Processing > 302.AI > LangExtract [Create information extraction task](https://doc-en.302.ai/337808547e0.md): LangExtract, an open-source project from Google: https://github.com/google/langextract - 根目录 > Information Processing > 302.AI > LangExtract [View information extraction task](https://doc-en.302.ai/337808548e0.md): View Information Extraction Task - 根目录 > Information Processing > 302.AI > Dots.OCR [Create dots.ocr Task](https://doc-en.302.ai/340677709e0.md): Create dots_ocr Task - 根目录 > Information Processing > 302.AI > Dots.OCR [View dots.ocr Task](https://doc-en.302.ai/340677710e0.md): View dots_ocr Task - 根目录 > Information Processing > 302.AI > MiniCPM [Create MiniCPM-V 4.5 Task](https://doc-en.302.ai/346048833e0.md): Create MiniCPM-V 4.5 tasks, enabling video analysis and document analysis. - 根目录 > Information Processing > 302.AI > MiniCPM [View MiniCPM-V 4.5 Task](https://doc-en.302.ai/346048834e0.md): View MiniCPM-V 4.5 Tasks - 根目录 > Information Processing > 302.AI > PDF Translation [Submit PDF Translation Task](https://doc-en.302.ai/353410868e0.md): From the open-source project: https://github.com/PDFMathTranslate/PDFMathTranslate-next - 根目录 > Information Processing > 302.AI > PDF Translation [Check PDF Translation Task](https://doc-en.302.ai/353410869e0.md): **Price: Free** - 根目录 > Information Processing > Tavily [Search](https://doc-en.302.ai/207705253e0.md): Tavily is a company focused on AI search. Their search is optimized for LLMs (Large Language Models) to facilitate data retrieval for them. - 根目录 > Information Processing > Tavily [Extract](https://doc-en.302.ai/235295291e0.md): Tavily is a company focused on AI search. Their search is optimized for LLMs (Large Language Models) to facilitate data retrieval for them. - 根目录 > Information Processing > SerpApi [Search](https://doc-en.302.ai/321674574e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（News）](https://doc-en.302.ai/321674575e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Images）](https://doc-en.302.ai/321674576e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Lens）](https://doc-en.302.ai/321674578e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Videos）](https://doc-en.302.ai/321674579e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Scholar）](https://doc-en.302.ai/321674581e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Patents）](https://doc-en.302.ai/321674582e0.md): SerpApi is a company that provides search APIs, allowing for quick and easy access to content from the Google search engine. We are fully aligned with their official interface, so you only need to replace the API Base URL. - 根目录 > Information Processing > SerpApi [Search（Baidu）](https://doc-en.302.ai/321741998e0.md): SerpApi is a company that provides a search API, allowing you to conveniently and quickly access content from the Google search engine. We are fully aligned with their official interface; you only need to replace the API Base URL. - 根目录 > Information Processing > Search1API [Search](https://doc-en.302.ai/235326268e0.md): Search1API is a company focused on search, and their distinctive feature is affordable pricing. - 根目录 > Information Processing > Search1API [News](https://doc-en.302.ai/235326270e0.md): Search1API is a company focused on search, and their distinctive feature is affordable pricing. - 根目录 > Information Processing > Search1API [Crawl](https://doc-en.302.ai/235326269e0.md): Search1API is a company focused on search, and their distinctive feature is affordable pricing. - 根目录 > Information Processing > Search1API [Sitemap（Site Map）](https://doc-en.302.ai/269263445e0.md): Search1API is a company focused on search services, and their main feature is affordability. - 根目录 > Information Processing > Search1API [Trending (Popular Trends)](https://doc-en.302.ai/276249952e0.md): Search1API is a company specializing in search, with their standout feature being affordable pricing. - 根目录 > Information Processing > Exa [Search](https://doc-en.302.ai/207705275e0.md): **Exa AI**, an emerging AI search engine company, recently announced that it has raised $17 million in Series A funding, led by Lightspeed with participation from Nvidia's NVentures and Y Combinator. Unlike other search engines, Exa aims to become a dedicated search tool for AI. - 根目录 > Information Processing > Exa [Contents（Get content）](https://doc-en.302.ai/207705276e0.md): **Exa AI**, an emerging AI search engine company, recently announced that it has raised $17 million in Series A funding, led by Lightspeed with participation from Nvidia's NVentures and Y Combinator. Unlike other search engines, Exa aims to become a dedicated search tool for AI. - 根目录 > Information Processing > Exa [Answer](https://doc-en.302.ai/276258454e0.md): Exa AI is an emerging AI search engine company that recently announced a $17 million Series A funding round, led by Lightspeed, with participation from Nvidia's NVentures and Y Combinator. Unlike other search engines, Exa aims to be a dedicated search tool for AI. - 根目录 > Information Processing > Bocha AI [Web-search](https://doc-en.302.ai/207705277e0.md): Invoke AI search to answer user questions, returning multimodal reference sources, summarized answers, and follow-up questions. Reference sources (web pages, TikTok videos, images), summarized answers, follow-up questions. - 根目录 > Information Processing > Bocha AI [Ai-search](https://doc-en.302.ai/207705278e0.md): Utilize AI search to answer user questions, providing multimodal reference sources, summarized answers, and follow-up questions. Reference sources (web pages, TikTok videos, images), summarized answers, follow-up questions. - 根目录 > Information Processing > Doc2x > Version 2 [PDF（Upload - Asynchronous）](https://doc-en.302.ai/232502061e0.md): Upload the PDF to start parsing. - 根目录 > Information Processing > Doc2x > Version 2 [Status（View Status）](https://doc-en.302.ai/232502063e0.md): Check the processing status after uploading the PDF. - 根目录 > Information Processing > Doc2x > Version 2 [Parse（Request Export File - Asynchronous）](https://doc-en.302.ai/232502062e0.md): Export the uploaded PDFs to other formats. - 根目录 > Information Processing > Doc2x > Version 2 [Result（exported results） ](https://doc-en.302.ai/232502064e0.md): Getting the exported results - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [PDF（PDF-to-MD）](https://doc-en.302.ai/207705261e0.md): **Convert PDFs to MD format**, from our partner **Doc2x**: https://doc2x.com/ - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [PDF-Async](https://doc-en.302.ai/207705264e0.md): **Convert PDFs to any format**, from our partner **Doc2x**: https://doc2x.com/ - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [IMG-to-MD](https://doc-en.302.ai/207705262e0.md): **Convert IMG to MD format**, from our partner **Doc2x**: https://doc2x.com/ - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [IMG-Async](https://doc-en.302.ai/207705263e0.md): **Convert IMG to any format**, from our partner **Doc2x**: https://doc2x.com/ - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [Status](https://doc-en.302.ai/207705265e0.md): Get conversion status. - 根目录 > Information Processing > Doc2x > Version 1 (Deprecated) [Export](https://doc-en.302.ai/207705266e0.md): Export file, optional formats: - 根目录 > Information Processing > Glif [Glif（Bot）](https://doc-en.302.ai/207705267e0.md): **Glif is a bot-building platform similar to Coze**: https://glif.app/ - 根目录 > Information Processing > Jina [Reader（Web Page to Markdown）](https://doc-en.302.ai/207705268e0.md): Feeding web information into LLMs is an important step for grounding, but it can be challenging. The simplest approach is to scrape web pages and provide the raw HTML. However, scraping can be complex and is often blocked, and the raw HTML is filled with extraneous elements like tags and scripts. Reader API addresses these issues by extracting the core content from the URL and converting it into clean, LLM-friendly text, ensuring that your agents and RAG systems receive high-quality input. - 根目录 > Information Processing > Jina [Search](https://doc-en.302.ai/207705269e0.md): LLMs have a knowledge cutoff point, which means they cannot access the most up-to-date world knowledge. This can lead to issues such as misinformation, outdated responses, hallucinations, and other factual errors. For GenAI applications, grounding is absolutely essential. Reader allows you to ground your LLM with the latest information from the web. Simply prepend your query with https://s.jina.ai/, and Reader will search the web and return the top five results along with their URLs and content, each formatted as clean, LLM-friendly text. This way, you can keep your LLM up-to-date, improve its accuracy, and reduce hallucinations. - 根目录 > Information Processing > Jina [Grounding（Verification of Facts）](https://doc-en.302.ai/230624831e0.md): Fact-checking by searching, from Jina - 根目录 > Information Processing > Jina [Classify](https://doc-en.302.ai/230629283e0.md): Categorizing and tagging content using embedding models，from：https://jina.ai/classifier - 根目录 > Information Processing > DeepL [Chat（Translate into English）](https://doc-en.302.ai/207705272e0.md): Translation service from **DeepL** - 根目录 > Information Processing > DeepL [Chat（Translate into Chinese）](https://doc-en.302.ai/207705273e0.md): Translation service from **DeepL** - 根目录 > Information Processing > DeepL [Chat（Translate into Japanese）](https://doc-en.302.ai/207705274e0.md): Translation service from **DeepL** - 根目录 > Information Processing > DeepL [Translate（Translate into various language）](https://doc-en.302.ai/214730964e0.md): Translation service from DeepL - 根目录 > Information Processing > RSSHub [RSSHub](https://doc-en.302.ai/235940255e0.md): RSSHub is an open-source project that converts a wide variety of websites into RSS format data, making it convenient for users to receive updates in a timely manner, including but not limited to public accounts and Xiaohongshu. - 根目录 > Information Processing > Firefly card [saveImg（Card Generation）](https://doc-en.302.ai/245142413e0.md): On the webpage https://fireflycard.shushiai.com/ editing the card,click to copy the JSON and paste it as a body parameter. - 根目录 > Information Processing > Youdao [Youdao（Youdao Translate）](https://doc-en.302.ai/250800740e0.md): Translation API from Youdao - 根目录 > Information Processing > Mistral [OCR（PDF Parsing)](https://doc-en.302.ai/270148035e0.md): PDF parsing by Mistral allows you to quickly convert PDFs into MD. - 根目录 > Information Processing > Firecrawl [Scrape](https://doc-en.302.ai/288419862e0.md): A web scraping service from Firecrawl that can convert web pages into Markdown format. - 根目录 > Information Processing > Firecrawl [Batch Scrape](https://doc-en.302.ai/288419926e0.md): A web scraping service from Firecrawl that can convert web pages into Markdown (md) format - 根目录 > Information Processing > Firecrawl [Get Batch Scrape Status](https://doc-en.302.ai/288419983e0.md): A web scraping service from Firecrawl that can convert web pages into Markdown (md) format - 根目录 > Information Processing > Firecrawl [Get Batch Scrape Errors](https://doc-en.302.ai/288420166e0.md): A web scraping service from Firecrawl that can convert web pages into Markdown format - 根目录 > Information Processing > Firecrawl [Map](https://doc-en.302.ai/288420246e0.md): Web sitemap retrieval service from Firecrawl - 根目录 > Information Processing > Firecrawl [Search](https://doc-en.302.ai/288996528e0.md): Search service from Firecrwal - 根目录 > Information Processing > MetaSota Search [Search](https://doc-en.302.ai/327572457e0.md): Return real-time search results from web and other data sources based on questions or keywords - 根目录 > Information Processing > MetaSota Search [Reader](https://doc-en.302.ai/327572458e0.md): Fetch the full content of the corresponding webpage based on the input URL - 根目录 > Information Processing > MetaSota Search [Chat](https://doc-en.302.ai/327572459e0.md): Return the corresponding answer based on the query - 根目录 > Information Processing > MinerU [Create PDF Parsing Task (Open Source Deployment Version)](https://doc-en.302.ai/331998206e0.md): Open-source V2 version deployed at 302, create PDF parsing task - 根目录 > Information Processing > MinerU [View PDF Parsing Task (Open Source Deployment Version)](https://doc-en.302.ai/331998207e0.md): View PDF Parsing Task - 根目录 > Information Processing > MinerU [Create PDF Parsing Task (Official Free Version)](https://doc-en.302.ai/331998208e0.md): Reference: https://mineru.net/apiManage/docs - 根目录 > Information Processing > MinerU [View PDF Parsing Task (Official Free Version)](https://doc-en.302.ai/331998209e0.md): Reference: https://mineru.net/apiManage/docs - 根目录 > Information Processing > Zhipu Agent [Zhipu PPT Creation](https://doc-en.302.ai/340677711e0.md): GLM PPT is a next-generation intelligent tool designed for professionals and creators. Powered by the GLM large model, it differs from traditional engineering-based assembly solutions by enabling one-click generation of interactive slides from natural language instructions. Deeply integrating content generation with design standards, it can quickly deliver professional-grade works, lowering the design threshold and improving content production efficiency. - 根目录 > Information Processing > Zhipu Agent [Zhipu PPT Export](https://doc-en.302.ai/340677712e0.md): Export PDF - 根目录 > Information Processing > Unifuncs [Web-Search (Real-time Search)](https://doc-en.302.ai/344140561e0.md): **Price: 0.01 PTC / time** - 根目录 > Information Processing > Unifuncs [Web-Reader (Webpage Reading)](https://doc-en.302.ai/344140562e0.md): **Price: 0.015 PTC / time** - 根目录 > Information Processing > Sophnet [Image Recognition](https://doc-en.302.ai/355542344e0.md): From Spohnet: Image OCR - 根目录 > Information Processing > Sophnet [Document Recognition](https://doc-en.302.ai/355542345e0.md): From Sophnet Document Recognition - 根目录 > Information Processing > Doubao [Doubao-Seed-Translation](https://doc-en.302.ai/357382287e0.md): Doubao-Seed-Translation is a self-developed multilingual translation model supporting translation between 28 languages. It delivers highly accurate, natural, and fluent translations. Its Chinese-English translation quality is comparable to Deepseek-R1, and its general multilingual translation performance matches or surpasses GPT-4o / Gemini-2.5-Pro. The model is precisely adapted for various scenarios, including office and entertainment. It supports a 4K context window and can output up to 3K tokens. - 根目录 > Information Processing > Perplexity [Search](https://doc-en.302.ai/357382286e0.md): From Perplexity Search API - 根目录 > Information Processing > Aminer > Academic Q&A [Paper Q&A Search](https://doc-en.302.ai/357688191e0.md): **Price: 0.008 PTC/time** - 根目录 > Information Processing > Aminer > Academic Q&A [AMiner Meditation](https://doc-en.302.ai/357688192e0.md): **Price: 0.08 PTC/time** - 根目录 > Information Processing > Aminer > Data Acquisition [Journal Papers](https://doc-en.302.ai/357751125e0.md): **价格：0.02 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Institution Details](https://doc-en.302.ai/357751126e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Patent Details](https://doc-en.302.ai/357751127e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Paper Details](https://doc-en.302.ai/357751128e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [ Scholar Portrait](https://doc-en.302.ai/357751129e0.md): **价格：0.08 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Scholar Details](https://doc-en.302.ai/357751130e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Paper Citations](https://doc-en.302.ai/357751131e0.md): **价格：0.02 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [ Scholar Papers](https://doc-en.302.ai/357751132e0.md): **价格：0.25 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Scholar Patents](https://doc-en.302.ai/357751133e0.md): **价格：0.25 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [ Paper Information](https://doc-en.302.ai/357751134e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [ Patent Information](https://doc-en.302.ai/357751135e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Institution Papers](https://doc-en.302.ai/357751136e0.md): **价格：0.02 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [ Institution Scholars](https://doc-en.302.ai/357751137e0.md): **价格：0.08 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Institution Patents](https://doc-en.302.ai/357751138e0.md): **价格：0.02 PTC/次** - 根目录 > Information Processing > Aminer > Data Acquisition [Journal Details](https://doc-en.302.ai/357751139e0.md): **价格：0.04 PTC/次** - 根目录 > Information Processing > Aminer > Data Disambiguation [Institution Disambiguation Pro](https://doc-en.302.ai/357751140e0.md): **价格：0.008 PTC/次** - 根目录 > Information Processing > Aminer > Data Disambiguation [Institution Disambiguation](https://doc-en.302.ai/357751141e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [Journal Search](https://doc-en.302.ai/357751142e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [Institution Search](https://doc-en.302.ai/357751143e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [ Patent Search](https://doc-en.302.ai/357751144e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [Paper Search](https://doc-en.302.ai/357751145e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [Paper Search Pro](https://doc-en.302.ai/357751146e0.md): **价格：0.002 PTC/次** - 根目录 > Information Processing > Aminer > Data Query [Scholar Search](https://doc-en.302.ai/357751147e0.md): **价格：0 PTC/次** - 根目录 > Information Processing > Aminer > Composite Interface [Get Paper Details by Conditions](https://doc-en.302.ai/357751148e0.md): **价格：0.04 PTC/次** - 根目录 > Information Processing > Aminer > Composite Interface [ Paper Search API](https://doc-en.302.ai/357751149e0.md): **价格：0.05 PTC/次** - 根目录 > Information Processing > Aminer > Composite Interface [Paper Batch Query API](https://doc-en.302.ai/357751150e0.md): **价格：0.02 PTC/次** - 根目录 > RAG-related > OpenAI [Embeddings](https://doc-en.302.ai/207705279e0.md): [OpenAI Guide](https://platform.openai.com/docs/guides/text-generation/chat-completions-api) - 根目录 > RAG-related > Jina [Embeddings](https://doc-en.302.ai/207705286e0.md): Embedding model from Jina: https://jina.ai/embeddings - 根目录 > RAG-related > Jina [Rerank](https://doc-en.302.ai/207705288e0.md): From Jina's rerank model, it can greatly improve retrieval accuracy in RAG. - 根目录 > RAG-related > Jina [Rerank（Multimodal Reordering）](https://doc-en.302.ai/282201344e0.md): Jina's multimodal rerank model can simultaneously rank images and text. - 根目录 > RAG-related > Jina [Tokenizer](https://doc-en.302.ai/207705287e0.md): Tokenize model from Jina: https://jina.ai/tokenizer/ - 根目录 > RAG-related > China Model [Embeddings（Zhipu） ](https://doc-en.302.ai/207705280e0.md): embeddings用于衡量文本字符串之间的相关性。embeddings通常用于： - 根目录 > RAG-related > China Model [Embeddings（BAAI）](https://doc-en.302.ai/207705285e0.md): Support two models： - 根目录 > RAG-related > China Model [Embeddings（Baichuan AI） ](https://doc-en.302.ai/207705281e0.md): Text embeddings are used to measure the relevance between text strings. Embeddings are typically used for: - 根目录 > RAG-related > China Model [Embeddings（Youdao） ](https://doc-en.302.ai/207705282e0.md): Text embeddings are used to measure the relevance between text strings. Embeddings are typically used for: - 根目录 > RAG-related > China Model [Rerank（Youdao） ](https://doc-en.302.ai/207705283e0.md): **Price：0.02 PTC / 1M tokens** - 根目录 > RAG-related > China Model [Rerank（BAAI）](https://doc-en.302.ai/207705284e0.md): **Price：0.02 PTC / 1M tokens** - 根目录 > RAG-related > 302.AI [Chat（with KB）](https://doc-en.302.ai/222611742e0.md): For Knowledge Base conversations, the API Key of the Knowledge Base robot will be automatically bound to the Knowledge Base selected in the background, and kb_id does not need to be passed when using this type of key - 根目录 > RAG-related > 302.AI [Chat（with KB-OpenAI compatible）](https://doc-en.302.ai/238962472e0.md): API Key needs to be filled with the key corresponding to the Knowledge Base robot - 根目录 > RAG-related > 302.AI [Create（Knowledge Base）](https://doc-en.302.ai/222611737e0.md): Create a Knowledge Base - 根目录 > RAG-related > 302.AI [Delete（Knowledge Base）](https://doc-en.302.ai/222611741e0.md): Delete the specified knowledge base - 根目录 > RAG-related > 302.AI [Upload](https://doc-en.302.ai/222611738e0.md): Uploading files to a specified knowledge base - 根目录 > RAG-related > 302.AI [Delete (Delete the specified file from the Knowledge Base)](https://doc-en.302.ai/404764904e0.md): Delete the specified Knowledge Base - 根目录 > RAG-related > 302.AI [List（KB）](https://doc-en.302.ai/222611739e0.md): Get a list of knowledge bases - 根目录 > RAG-related > 302.AI [Info](https://doc-en.302.ai/222611740e0.md): Access to Knowledge Base Details - 根目录 > RAG-related > 302.AI [Meta-Chunking（Text LLM slices）](https://doc-en.302.ai/245142513e0.md): Use LLM's text comprehension to segment content, ensuring the slices maintain coherent context - 根目录 > RAG-related > 302.AI [Meta-Chunking（File LLM slices）](https://doc-en.302.ai/245142548e0.md): Use LLM's text comprehension to segment content, ensuring the slices maintain coherent context - 根目录 > RAG-related > SiliconFlow [Embeddings](https://doc-en.302.ai/323761393e0.md): Embeddings Models from Silicon Flow - 根目录 > RAG-related > SiliconFlow [Rerank](https://doc-en.302.ai/323761394e0.md): The Rerank model from SiliconFlow can greatly improve recall accuracy in RAG. - 根目录 > RAG-related > Google [Embeddings](https://doc-en.302.ai/323758845e0.md): Embedding Model from Google - 根目录 > RAG-related > Voyoge [Embeddings](https://doc-en.302.ai/369532877e0.md): Vector Generation from Voyoga - 根目录 > RAG-related > Voyoge [ContextualizedEmbeddings](https://doc-en.302.ai/404772645e0.md): Voyage's context-aware block embedding endpoint accepts document blocks in addition to queries and full documents, and returns a response containing context-aware block vector embeddings. These context-aware block embeddings not only capture local details within each block but also capture global, Coarse Grain metadata from the entire document. - 根目录 > RAG-related > Voyoge [Rerank](https://doc-en.302.ai/369532878e0.md): The Rerank model from Voyoga can greatly improve recall accuracy in RAG. - 根目录 > Tools API > Pay with 302 [Create a payment order](https://doc-en.302.ai/376949056e0.md): Access steps: - 根目录 > Tools API > Pay with 302 [Get order status](https://doc-en.302.ai/376949057e0.md): Request order details via checkout_id - 根目录 > Tools API > AI Video Creation Hub [Scripts（Generate Video Content Copy）](https://doc-en.302.ai/248602281e0.md): The API used by the AI video content idea platform: Generate corresponding video scripts based on input keywords. - 根目录 > Tools API > AI Video Creation Hub [Terms（Generate Video Material Search Keywords）](https://doc-en.302.ai/248602282e0.md): The API used by the AI video content idea platform: Generate search keywords for video platforms based on the video theme and script. - 根目录 > Tools API > AI Video Creation Hub [Videos（Create Video Material Generation Task）](https://doc-en.302.ai/248602283e0.md): The API used by the AI video content idea platform: Generate stitched video materials based on the script and search keywords. - 根目录 > Tools API > AI Video Creation Hub [Tasks（Get Video Task Progress）](https://doc-en.302.ai/248602284e0.md): The API used by the AI video content idea platform: Retrieve the progress of a video generation task - 根目录 > Tools API > AI Paper Writing > CO-STORM [Create generate article task](https://doc-en.302.ai/258284198e0.md): Enter topics and concerns to create a generated article task. - 根目录 > Tools API > AI Paper Writing > CO-STORM [Continue to generate dialogue interfaces](https://doc-en.302.ai/258284199e0.md): Simulation dialogue enables information acquisition. Users can enter specified content or let AI continue to generate simulated dialogue based on the above, thereby obtaining more information to generate articles. - 根目录 > Tools API > AI Paper Writing > CO-STORM [Update article content interface](https://doc-en.302.ai/258284200e0.md): After the call continues to generate the dialogue interface, the knowledge base corresponding to task_id will be updated. You can update the article through this interface. - 根目录 > Tools API > AI Paper Writing > CO-STORM [Get article information](https://doc-en.302.ai/258284201e0.md): **Price：0PTC/call** - 根目录 > Tools API > AI Paper Writing [Asynchronous Paper Generate](https://doc-en.302.ai/245500983e0.md): Asynchronous API for Generating Papers in an AI Writing Tool - 根目录 > Tools API > AI Paper Writing [Fetch](https://doc-en.302.ai/245500984e0.md): Fetch Paper Results - 根目录 > Tools API > AI Podcast Production [Asynchronous Generate Podcast Transcripts](https://doc-en.302.ai/245500985e0.md): Interfaces used in AI podcast generation tools. - 根目录 > Tools API > AI Podcast Production [Check the status of podcast text generation task](https://doc-en.302.ai/245500986e0.md): Check task status - 根目录 > Tools API > AI Podcast Production [Asynchronously Generate Podcast Audio](https://doc-en.302.ai/245500987e0.md): Interfaces used in AI podcast generation tools. - 根目录 > Tools API > AI Podcast Production [Check the status of podcast audio generation task](https://doc-en.302.ai/245500988e0.md): Check task status - 根目录 > Tools API > AI Podcast Production [Get supported timbre information](https://doc-en.302.ai/404383117e0.md): - 根目录 > Tools API > AI Writing Assistant [Get Tools‘ List](https://doc-en.302.ai/246101919e0.md): Get a list of copywriting tools, primarily used for retrieving tool information and the parameter attributes for generating copywriting through APIs. - 根目录 > Tools API > AI Writing Assistant [Generate Copywriting](https://doc-en.302.ai/246101920e0.md): Retrieve the corresponding `tool_name` and `params` from the interface for obtaining the copywriting tool list, and pass them to this interface to generate copywriting. - 根目录 > Tools API > AI Video Real-Time Translation [Query Video Information](https://doc-en.302.ai/247577932e0.md): Retrieve video title, thumbnail, resolution format list, and playback link. - 根目录 > Tools API > AI Video Real-Time Translation [Video Download](https://doc-en.302.ai/247577933e0.md): Download YouTube and Bilibili videos, upload them to a 302 file server, and extract the audio track. - 根目录 > Tools API > AI Video Real-Time Translation [Extract Audio from Video](https://doc-en.302.ai/248490188e0.md): Extract Audio from Video - 根目录 > Tools API > AI Video Real-Time Translation [Audio vocal separation and transcription](https://doc-en.302.ai/255892113e0.md): Use our own optimized whisper model to transcribe audio into word-level text data - 根目录 > Tools API > AI Video Real-Time Translation [Subtitle Translation](https://doc-en.302.ai/247577935e0.md): Input word-level audio transcription results and return translated SRT subtitles. - 根目录 > Tools API > AI Video Real-Time Translation [Video Burning](https://doc-en.302.ai/247577936e0.md): Burn SRT subtitles into the video. - 根目录 > Tools API > AI Video Real-Time Translation [Original sound clone](https://doc-en.302.ai/255891913e0.md): Output new voice according to the original voice and new audio content, suitable for dubbing audio in different languages - 根目录 > Tools API > AI Video Real-Time Translation [Query task status](https://doc-en.302.ai/247577937e0.md): Query task status - 根目录 > Tools API > AI Video Real-Time Translation [Get the specified frame of the video link](https://doc-en.302.ai/404408355e0.md): Under Development - 根目录 > Tools API > AI Document Editor [Generate a long text outline](https://doc-en.302.ai/258267631e0.md): **Price: Charge based on the calling model** - 根目录 > Tools API > AI Document Editor [Generate article content](https://doc-en.302.ai/258267632e0.md): Streaming response - 根目录 > Tools API > Web Data Extraction Tool [Generate Schema](https://doc-en.302.ai/261542258e0.md): Generate corresponding Schema through web page and description - 根目录 > Tools API > Web Data Extraction Tool [Create an extraction task](https://doc-en.302.ai/261542259e0.md): Create a web crawling task - 根目录 > Tools API > Web Data Extraction Tool [Query extraction progress](https://doc-en.302.ai/261542260e0.md): Get the progress of web crawling tasks - 根目录 > Tools API > AI Prompt Expert [Prompt Optimization](https://doc-en.302.ai/280482594e0.md): Interface responses are consistent with OpenAI chat model responses, supporting streaming returns. - 根目录 > Tools API > AI Prompt Expert [Image prompt generation](https://doc-en.302.ai/261664608e0.md): Convert a picture into an AI drawing prompt word that is used to generate a picture - 根目录 > Tools API > AI Prompt Expert [Create SPO Prompt Optimization Task](https://doc-en.302.ai/279192337e0.md): Example Usage: - 根目录 > Tools API > AI Prompt Expert [Query SPO Prompt Optimization Results](https://doc-en.302.ai/279192443e0.md): **Price： 0 PTC/call** - 根目录 > Tools API > AI 3D Modeling [3D model file type conversion](https://doc-en.302.ai/261680304e0.md): From open source project: https://github.com/mikedh/trimesh - 根目录 > Tools API > AI Search Master 3.0 [AI Search](https://doc-en.302.ai/265291523e0.md): The search API in Search Master only supports streaming mode and needs to be adapted by yourself. - 根目录 > Tools API > AI Vector Graphics Generation [SVG to video](https://doc-en.302.ai/265298816e0.md): Convert SVG to video of the generation process. - 根目录 > Tools API > Al Answer Machine [Answer](https://doc-en.302.ai/261663767e0.md): Support JSON parameters (content image link or BS64 images (starting from data:image) or title text) and FormData parameters (content binary images) - 根目录 > Tools API > AI PPT Generator [Generate PPT interface with one click](https://doc-en.302.ai/265304598e0.md): ** Price: 0.07PTC/call ** - 根目录 > Tools API > AI PPT Generator [File parsing](https://doc-en.302.ai/265304599e0.md): The generated file link is only effective on the same day - 根目录 > Tools API > AI PPT Generator [Generate an outline](https://doc-en.302.ai/265304600e0.md): **Price: Free** - 根目录 > Tools API > AI PPT Generator [Generate outline content](https://doc-en.302.ai/265304601e0.md): When selecting synchronously to generate PPT, there is no deduction for this interface. Only when synchronously generate PPT interface will the deduction trigger. - 根目录 > Tools API > AI PPT Generator [Get template options](https://doc-en.302.ai/265304602e0.md): **Price: Free** - 根目录 > Tools API > AI PPT Generator [Generate PPT interface (synchronous interface)](https://doc-en.302.ai/265304603e0.md): **Price: 0.07PTC/call** - 根目录 > Tools API > AI PPT Generator [Load PPT data](https://doc-en.302.ai/265304604e0.md): **Price: Free** - 根目录 > Tools API > AI PPT Generator [Generate PPT interface (asynchronous interface)](https://doc-en.302.ai/265304605e0.md): When asyncGenPptx=true is requested to generate outline content, PPT will be generated asynchronously. At this time, there is no need to call the generated PPT interface again. - 根目录 > Tools API > AI PPT Generator [Asynchronous query generates PPT status](https://doc-en.302.ai/265304606e0.md): Note: This interface can only query data (temporary cache data) during streaming generation. The data will fail after 30 seconds of the response. - 根目录 > Tools API > AI PPT Generator [Download PPT](https://doc-en.302.ai/265304607e0.md): **Price: Free** - 根目录 > Tools API > AI PPT Generator [Add/update custom PPT templates](https://doc-en.302.ai/265304608e0.md): The uploaded template will be isolated according to the apikey. When querying the custom template, only the template data uploaded by the corresponding apikey will be returned. - 根目录 > Tools API > AI PPT Generator [Pagination query PPT template](https://doc-en.302.ai/265304609e0.md): **Price: Free** - 根目录 > Tools API > AI Academic Paper Search [arxiv Paper Search](https://doc-en.302.ai/265250453e0.md): Search for arxiv paper and translate title - 根目录 > Tools API > AI Academic Paper Search [Google Paper Search](https://doc-en.302.ai/265250454e0.md): **Price: 0.005PTC/call, no charge for hitting cache** - 根目录 > Tools API > One-Click Website Deployment [Create hosted webpage（github）](https://doc-en.302.ai/363182866e0.md): Supports deploying full-stack web pages directly by providing a GitHub link. - 根目录 > Tools API > One-Click Website Deployment [Create hosted webpage (Example of Passing Parameters via form-data)](https://doc-en.302.ai/404617562e0.md): Supports passing in a GitHub link to directly deploy a full stack web page. - 根目录 > Tools API > One-Click Website Deployment [Query hosted project details](https://doc-en.302.ai/363182867e0.md): Compatible website one-click deployment for all interface-generated web project IDs - 根目录 > Tools API > One-Click Website Deployment [Create hosted webpage（json） ](https://doc-en.302.ai/285348343e0.md): **Pricing: Create hosted webpage: 0.001 PTC/time, Generate webpage cover image (optional service) 0.001 PTC/1 time** - 根目录 > Tools API > One-Click Website Deployment [Create hosted webpage（form-data）](https://doc-en.302.ai/285347675e0.md): Note: If the uploaded code contains syntax issues, the API will not perform code validation (including bugs such as font garbling caused by unspecified encoding) and will proceed with normal deployment. - 根目录 > Tools API > One-Click Website Deployment [Create hosted webpage (binary)](https://doc-en.302.ai/285348409e0.md): Place the html file or md file directly in the request body in binary form, and put other parameters in the query. - 根目录 > Tools API > One-Click Website Deployment [Query hosted project list](https://doc-en.302.ai/285348505e0.md): Only hosted projects under the current API Key can be queried - 根目录 > Tools API > One-Click Website Deployment [Delete deployed webpage](https://doc-en.302.ai/363182868e0.md): - 根目录 > Tools API > AI Avatar Maker [Generate Avatar (JSON Parameter Example)](https://doc-en.302.ai/295534159e0.md): **Price：Optimize the interface cost of tokens+avatars generated by prompts(`/302/submit/flux-selfie`)** - 根目录 > Tools API > AI Avatar Maker [Avatar Generation (form-data parameter example)](https://doc-en.302.ai/295534010e0.md): **Price：Optimize the interface cost of tokens+avatars generated by prompts(`/302/submit/flux-selfie`)** - 根目录 > Tools API > AI Card Generation [Generate Knowledge Card](https://doc-en.302.ai/297347900e0.md): You can view the specific styles at: https://302.ai/product/detail/ai-card-generation - 根目录 > Tools API > AI Card Generation [Generate Poster](https://doc-en.302.ai/297347933e0.md): You can view the specific styles at: https://302.ai/product/detail/ai-card-generation - 根目录 > Tools API > AI Card Generation [Generate philosophical cards](https://doc-en.302.ai/297347962e0.md): You can view the specific styles at: https://302.ai/product/detail/ai-card-generation - 根目录 > Tools API > AI Card Generation [Generate philosophical quotation sayings](https://doc-en.302.ai/297347996e0.md): You can view the specific styles at: https://302.ai/product/detail/ai-card-generation - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Basic Text-to-Image (Optional Prompt Optimization)](https://doc-en.302.ai/346924433e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/1 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Style Modification](https://doc-en.302.ai/346924434e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/2 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [English Word Flashcards](https://doc-en.302.ai/346924435e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/3 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Visual Recipe](https://doc-en.302.ai/346924436e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/4 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Physical Destruction Effect Card](https://doc-en.302.ai/346924437e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/5 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Product Model Image](https://doc-en.302.ai/346924438e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/6 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Passport Stamp Generator](https://doc-en.302.ai/346924439e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/7 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Themed Keycap Scene Generation](https://doc-en.302.ai/346924440e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/8 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Low Polygon](https://doc-en.302.ai/346924441e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/9 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Typography](https://doc-en.302.ai/346924442e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/10 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Sculpture Generation](https://doc-en.302.ai/346924443e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/11 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Isometric Miniature Scene Generation](https://doc-en.302.ai/346924444e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/12 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Unfolded Ancient Book Miniature Scene Generation](https://doc-en.302.ai/346924445e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/13 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [City Isometric View Generation](https://doc-en.302.ai/346924446e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/14 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Blister Tablet Food Generation](https://doc-en.302.ai/346924447e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/15 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Chibi-style 3D character creation](https://doc-en.302.ai/346924448e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/16 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D Relief Papercut Style Generation](https://doc-en.302.ai/346924449e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/17 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Change Character Age](https://doc-en.302.ai/346924450e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/18 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Movable Doll Generation](https://doc-en.302.ai/346924451e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/19 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D avatar pose generation](https://doc-en.302.ai/346924452e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/20 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Clay Style Generation](https://doc-en.302.ai/346924453e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/21 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D stereoscopic model creation](https://doc-en.302.ai/346924454e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/22 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Character Dual Exposure](https://doc-en.302.ai/346924455e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/23 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Hand-drawn Style Infographic Card](https://doc-en.302.ai/346924456e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/24 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Ghibli](https://doc-en.302.ai/346924457e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/25 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [LEGO Collection](https://doc-en.302.ai/346924458e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/26 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [CrystalBall](https://doc-en.302.ai/346924459e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/27 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Microscopic World](https://doc-en.302.ai/346924460e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/28 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Sticker Design](https://doc-en.302.ai/346924461e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/29 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Journal Notes](https://doc-en.302.ai/346924462e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/30 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Cloud Art](https://doc-en.302.ai/346924463e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/31 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Miniature 3D building](https://doc-en.302.ai/346924464e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/32 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Fictional Tweet Screenshot Prompt Optimization](https://doc-en.302.ai/346924465e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/33 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Cute Enamel Pin Image](https://doc-en.302.ai/346924466e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/34 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Ultra-Realistic Figurine Image](https://doc-en.302.ai/346924467e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/35 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Brand Pill Chart](https://doc-en.302.ai/346924468e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/36 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Original Product Image](https://doc-en.302.ai/346924469e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/37 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [AlphabetBox](https://doc-en.302.ai/346924470e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/38 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Plastic Garbage Bag](https://doc-en.302.ai/346924471e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/39 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D scroll miniature scene](https://doc-en.302.ai/346924472e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/40 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Silhouette Art](https://doc-en.302.ai/346924473e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/41 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Colorful Vector Art Poster](https://doc-en.302.ai/346924474e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/42 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Retro Promotional Poster](https://doc-en.302.ai/346924475e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/43 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Fashion Magazine Cover](https://doc-en.302.ai/346924476e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/44 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Mini Tilt-Shift Landscape](https://doc-en.302.ai/346924477e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/45 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [ Chibi-style keychain](https://doc-en.302.ai/346924478e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/46 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D miniature shop](https://doc-en.302.ai/346924479e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/47 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Nail Painting](https://doc-en.302.ai/346924480e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/48 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Pin on T-shirt](https://doc-en.302.ai/346924481e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/49 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [LEGO City Attractions](https://doc-en.302.ai/346924482e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/50 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Word and Graphic Fusion](https://doc-en.302.ai/346924483e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/51 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D chromed badge](https://doc-en.302.ai/346924484e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/52 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Animal Landmark Selfie](https://doc-en.302.ai/346924485e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/53 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Custom Anime Figure](https://doc-en.302.ai/346924486e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/54 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Frosted Glass Silhouette](https://doc-en.302.ai/346924487e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/55 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Rusty Iron Plate](https://doc-en.302.ai/346924488e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/56 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [FlowInk](https://doc-en.302.ai/346924489e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/57 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Neon Graffiti](https://doc-en.302.ai/346924490e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/58 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Claw Machine](https://doc-en.302.ai/346924491e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/59 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Creative Minimalist Ad](https://doc-en.302.ai/346924492e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/60 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Glass Shard](https://doc-en.302.ai/346924493e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/61 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Creative Drawstring Bag](https://doc-en.302.ai/346924494e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/62 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Floral Sculpture](https://doc-en.302.ai/346924495e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/63 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Succulent Planter](https://doc-en.302.ai/346924496e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/64 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Retro Sci-Fi Book Cover](https://doc-en.302.ai/346924497e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/65 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Gold Coin](https://doc-en.302.ai/346924498e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/66 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Emotion Pastry](https://doc-en.302.ai/346924499e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/67 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Fun Balloon](https://doc-en.302.ai/346924500e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/68 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [MonsterLetter](https://doc-en.302.ai/346924501e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/69 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [City in Toy Box](https://doc-en.302.ai/346924502e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/70 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [3D Doll](https://doc-en.302.ai/347297370e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/71 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Clothing Flat Lay](https://doc-en.302.ai/347297371e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/72 - 根目录 > Tools API > AI Image Creative Station API > Usage Example [Anime to Real Person](https://doc-en.302.ai/347297372e0.md): The request URL can also be changed to /302/gpt-image-creative/generate/73 - 根目录 > Tools API > AI Image Creative Station API [Get Model List](https://doc-en.302.ai/299580470e0.md): - 根目录 > Tools API > AI Image Creative Station API [Generate Creative Images](https://doc-en.302.ai/299580471e0.md): Since image generation takes a long time, please do not debug online in APIFOX. - 根目录 > Tools API > AI Image Creative Station API [Get Image Generation Result Asynchronously](https://doc-en.302.ai/346924432e0.md): - 根目录 > Tools API > AI Digital Human [Submit Video Processing Request](https://doc-en.302.ai/347009184e0.md): **Note: When extracting audio, if the input video lacks an audio track, a silent audio track of equal duration will be generated.** - 根目录 > Tools API > AI Digital Human [Query Audio Extraction Task Info](https://doc-en.302.ai/347009185e0.md): **Price: Query interface is free of charge** - 根目录 > Tools API > AI Digital Human [Query Video Merging Task Info](https://doc-en.302.ai/347009186e0.md): **Price: Query interface is free of charge** - 根目录 > Tools API > AI Digital Human [Video Merging Webhook Request Example](https://doc-en.302.ai/347009187e0.md): Please ensure that your webhook link is publicly accessible. - 根目录 > Tools API > AI Digital Human [Audio Extraction Webhook Request Example ](https://doc-en.302.ai/347009188e0.md): Please ensure that your webhook URL is publicly accessible. ## Schemas - 根目录 > 示例数据模型 [Pet](https://doc-en.302.ai/115969992d0.md): - 根目录 > 示例数据模型 [Category](https://doc-en.302.ai/115969993d0.md): - 根目录 > 示例数据模型 [Tag](https://doc-en.302.ai/115969994d0.md): - 根目录 > Schemas [Blend提交参数](https://doc-en.302.ai/115969995d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/175411397d0.md): - 根目录 > Schemas [任务](https://doc-en.302.ai/180232159d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/180346525d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/181868668d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/186871694d0.md): - 根目录 > Schemas [CreateLlmMrV2Schema](https://doc-en.302.ai/187694820d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/189708510d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/189711375d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/190456796d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/193103673d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/193740032d0.md): - 根目录 > Schemas [CreateLangExtractTaskSchema](https://doc-en.302.ai/194080110d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/194207860d0.md): - 根目录 > Schemas [创建点阵OCR任务结构](https://doc-en.302.ai/195847736d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/197047526d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/197086831d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/197207792d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/199197867d0.md): - 根目录 > Schemas [CheckActivationDTO](https://doc-en.302.ai/115969996d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/175411398d0.md): - 根目录 > Schemas [可执行按钮](https://doc-en.302.ai/180232160d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/180346526d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/181868669d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/186871695d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/187694821d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/189708511d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/189711376d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/190456797d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/193103674d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/193740033d0.md): - 根目录 > Schemas [CreateSmartVoiceTaskSchema](https://doc-en.302.ai/194207861d0.md): - 根目录 > Schemas [HTTPValidationError](https://doc-en.302.ai/195847737d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/197047527d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/197086832d0.md): - 根目录 > Schemas [验证错误](https://doc-en.302.ai/197207793d0.md): - 根目录 > Schemas [CreateMiniCPMV45ChatTaskSchema](https://doc-en.302.ai/199197868d0.md): - 根目录 > Schemas [Describe提交参数](https://doc-en.302.ai/115969997d0.md): - 根目录 > Schemas [提交结果](https://doc-en.302.ai/180232161d0.md): - 根目录 > Schemas [提交结果](https://doc-en.302.ai/180346527d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/187694822d0.md): - 根目录 > Schemas [CreateParsePdfTaskSchema](https://doc-en.302.ai/190456798d0.md): - 根目录 > Schemas [MergeImageSchema](https://doc-en.302.ai/193103675d0.md): - 根目录 > Schemas [LangExtractExampleSchema](https://doc-en.302.ai/194080111d0.md): - 根目录 > Schemas [CreateVoiceCloneTaskSchema](https://doc-en.302.ai/194207862d0.md): - 根目录 > Schemas [验证错误](https://doc-en.302.ai/195847738d0.md): - 根目录 > Schemas [ValidationError](https://doc-en.302.ai/199197869d0.md): - 根目录 > Schemas [Discord账号](https://doc-en.302.ai/115969998d0.md): - 根目录 > Schemas [LangExtractExampleExtractionSchema](https://doc-en.302.ai/194080112d0.md): - 根目录 > Schemas [验证错误](https://doc-en.302.ai/194207863d0.md): - 根目录 > Schemas [Imagine提交参数](https://doc-en.302.ai/115969999d0.md): - 根目录 > Schemas [验证错误](https://doc-en.302.ai/194080113d0.md): - 根目录 > Schemas [Map«string,object»](https://doc-en.302.ai/115970000d0.md): - 根目录 > Schemas [Message«Discord账号»](https://doc-en.302.ai/115970001d0.md): - 根目录 > Schemas [Message«Map«string,object»»](https://doc-en.302.ai/115970002d0.md): - 根目录 > Schemas [Message«Void»](https://doc-en.302.ai/115970003d0.md): - 根目录 > Schemas [Message«string»](https://doc-en.302.ai/115970004d0.md): - 根目录 > Schemas [Modal提交参数](https://doc-en.302.ai/115970005d0.md): - 根目录 > Schemas [Option](https://doc-en.302.ai/115970006d0.md): - 根目录 > Schemas [Pageable](https://doc-en.302.ai/115970007d0.md): - 根目录 > Schemas [Page«Map«string,object»»](https://doc-en.302.ai/115970008d0.md): - 根目录 > Schemas [Page«任务»](https://doc-en.302.ai/115970009d0.md): - 根目录 > Schemas [Shorten提交参数](https://doc-en.302.ai/115970010d0.md): - 根目录 > Schemas [Sort](https://doc-en.302.ai/115970011d0.md): - 根目录 > Schemas [TaskQueryDTO](https://doc-en.302.ai/115970012d0.md): - 根目录 > Schemas [任务](https://doc-en.302.ai/115970013d0.md): - 根目录 > Schemas [任务查询参数](https://doc-en.302.ai/115970014d0.md): - 根目录 > Schemas [变化任务提交参数](https://doc-en.302.ai/115970015d0.md): - 根目录 > Schemas [变化任务提交参数-simple](https://doc-en.302.ai/115970016d0.md): - 根目录 > Schemas [可执行按钮](https://doc-en.302.ai/115970017d0.md): - 根目录 > Schemas [可执行按钮_1](https://doc-en.302.ai/115970018d0.md): - 根目录 > Schemas [执行动作参数](https://doc-en.302.ai/115970019d0.md): - 根目录 > Schemas [提交结果](https://doc-en.302.ai/115970020d0.md): - 根目录 > Schemas [激活服务参数](https://doc-en.302.ai/115970021d0.md): - 根目录 > Schemas [账号修改参数](https://doc-en.302.ai/115970022d0.md): - 根目录 > Schemas [账号创建参数](https://doc-en.302.ai/115970023d0.md): - 根目录 > Schemas [账号查询参数](https://doc-en.302.ai/115970024d0.md):