userid
parameter.userid
appears for the first time, the backend will automatically generate a corresponding user and summarize the conversation through the LLM, extracting relevant information as long-term memory.userid
has appeared before, the backend will automatically retrieve the corresponding long-term memory and update the latest memory, automatically adding it to the system and passing it to the large model.userid
must comply with the uuidV4 standard; if it does not, the memory operation will not be executed. UUID generation test link: https://www.uuidgenerator.net/version4temperature
but not both.{
"model": "gpt-4o-mini",
"messages": [{"role": "user", "content": "我是谁?"}],
"userid":"fe657e7e-862e-4be0-990d-37428979a38c"
}
curl --location --request POST 'https://api.302.ai/v1/chat/completions' \
--header 'Accept: application/json' \
--header 'Authorization: Bearer sk-jls4AaVBGoe1GwZD64qZA1qyKTN1MPHa4NmvH1cT68z7K1Zz' \
--header 'Content-Type: application/json' \
--data-raw '{
"model": "gpt-4o-mini",
"messages": [{"role": "user", "content": "我是谁?"}],
"userid":"fe657e7e-862e-4be0-990d-37428979a38c"
}'
{
"id": "chatcmpl-123",
"object": "chat.completion",
"created": 1677652288,
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "\n\nHello there, how may I assist you today?"
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 9,
"completion_tokens": 12,
"total_tokens": 21
}
}