Hunyuan-Motion (motion generation)

POST

/302/submit/hunyuan-motion

302.AI's API comes from our own deployment model on cloud GPUs.Some models are open source models, and some models are fine-tuned or developed for ourselves.

A billion-parameter Vincentian 3D action model based on the Diffusion Transformer (DiT) model architecture and stream matching mechanism. It supports the generation of smooth and natural 3D character animations from natural language descriptions, covering a wide range of categories, and can be seamlessly integrated into the art 3D animation pipeline.
From: https://hunyuan.tencent.com/motion?tabIndex=0
Average run time 15-20s

** Price: 0.08 PTC/time **

Request

Header Params

Body Params application/json

Example

{
  "prompt": "Sit on the chair",
  "seed": 42,
  "duration": 3,
  "guidance_scale": 3,
  "output_format": "fbx"
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.302.ai/302/submit/hunyuan-motion' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
  "prompt": "Sit on the chair",
  "seed": 42,
  "duration": 3,
  "guidance_scale": 3,
  "output_format": "fbx"
}'

Responses

🟢200success

application/json

Body

Example

{
    "fbx_file": {
        "url": "https://file.302.ai/gpt/imgs/20260114/dbabd65a00bb09b0bdc8c7375c78fcb9.fbx",
        "content_type": "application/octet-stream",
        "file_size": 16768672
    },
    "motion_json": null,
    "seed": 42
}

Modified at 2026-01-14 10:33:29

Llava

Vectorize