思考模式(chat)

POST

/v1/chat/completions

大模型思考模式

请求参数

Header 参数

Body 参数application/json

示例

示例GLM示例

{
   "model": "gpt-5",
   "reasoning_effort": "high",
   "messages": [
      {
         "role": "user",
         "content": "hello"
      }
   ]
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

Curl

示例GLM示例

curl -X POST "https://api.wenwen-ai.com/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -d '{
        "model": "gpt-5",
        "reasoning_effort": "medium",
        "messages": [
          {
            "role": "user",
            "content": "讲个100字的故事"
          }
        ]
      }'

返回响应

🟢200成功

application/json

Body

示例

{
    "choices": [
        {
            "finish_reason": "stop",
            "index": 0,
            "logprobs": null,
            "message": {
                "annotations": [],
                "content": "村口老槐树忽然开灯般亮起，叶脉里流淌细小星火。阿禾摸着树皮，想起去世的奶奶。树影摇曳，仿佛一只温暖的手，把他推向回家的路。他快跑，门缝灯光像河，母亲端着热粥，笑里有泪。窗外的风停了。屋檐滴答，夜更深。",
                "refusal": null,
                "role": "assistant"
            }
        }
    ],
    "created": 1762789566,
    "id": "chatcmpl-CaONihwxVkSJ0jPmzbGaA02LcvWHa",
    "model": "gpt-5-2025-08-07",
    "object": "chat.completion",
    "system_fingerprint": null,
    "usage": {
        "completion_tokens": 1460,
        "completion_tokens_details": {
            "accepted_prediction_tokens": 0,
            "audio_tokens": 0,
            "reasoning_tokens": 1344,
            "rejected_prediction_tokens": 0
        },
        "prompt_tokens": 12,
        "prompt_tokens_details": {
            "audio_tokens": 0,
            "cached_tokens": 0
        },
        "total_tokens": 1472
    }
}

修改于 2026-04-07 05:57:57

流式返回(chat)

结构化输出(chat)