推理 API

旧版与已弃用

查看 Markdown

Completions(旧版)

/v1/completions

(旧版,推理模型不支持)根据给定提示创建文本补全响应。已由 /v1/chat/completions 替代。

请求体

响应体

choicesarray<object>

(array<object>,必填)— 模型的响应选项列表。长度对应请求体中的 `n`(默认为 1)。

createdinteger

(integer,必填)— chat completion 创建时间的 Unix 时间戳。

idstring

(string,必填)— 请求 ID。

modelstring

(string,必填)— 要使用的模型。

objectstring

(string,必填)— 响应对象类型,始终为 `"text_completion"`。

Exampletext

text

{
  "prompt": "1, 2, 3, 4, ",
  "model": "grok-3",
  "max_tokens": 3
}
Exampletext

text

{
  "id": "873492b3-6144-4279-ac2e-2c45242c5ce6",
  "object": "text_completion",
  "created": 1743771779,
  "model": "grok-3",
  "choices": [
    {
      "index": 0,
      "text": "5, ",
      "finish_reason": "length"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 3,
    "total_tokens": 15,
    "prompt_tokens_details": {
      "text_tokens": 12,
      "audio_tokens": 0,
      "image_tokens": 0,
      "cached_tokens": 0
    },
    "completion_tokens_details": {
      "reasoning_tokens": 0,
      "audio_tokens": 0,
      "accepted_prediction_tokens": 0,
      "rejected_prediction_tokens": 0
    }
  },
  "system_fingerprint": "fp_156d35dcaa"
}

Messages(兼容 Anthropic,已弃用)

/v1/messages

创建 messages 响应。此端点兼容 Anthropic API。

请求体

响应体

contentarray<object | object | object | object>

(array<object | object | object | object>,必填)— 响应消息内容。

idstring

(string,必填)— 唯一对象标识符。

modelstring

(string,必填)— 处理该请求的模型名称。

rolestring

(string,必填)— 生成消息的角色,始终为 `"assistant"`。

typestring

(string,必填)— 对象类型。对于 message 类型始终为 `"message"`。

usageobject

(object,必填)

Exampletext

text

{
  "model": "latest",
  "max_tokens": 32,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}
Exampletext

text

{
  "id": "4f224bfb-9d53-4c82-b40a-b7cd80831ec2",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello there! \"Hello, world\" is a classic, isn't it? Whether you're just saying hi or channeling your inner coder, I'm happy to greet you back"
    }
  ],
  "model": "latest",
  "stop_reason": "max_tokens",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 9,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "output_tokens": 32
  }
}

Completions(兼容 Anthropic,已弃用)

/v1/complete

(旧版 - 推理模型不支持)创建文本补全响应。此端点兼容 Anthropic API。

请求体

响应体

completionstring

(string,必填)— completion 内容,不包含停止序列。

idstring

(string,必填)— completion 响应的 ID。

modelstring

(string,必填)— 处理该请求的模型。

typestring

(string,必填)— completion 响应对象类型,始终为 `"completion"`。

Exampletext

text

{
  "model": "grok-3",
  "max_tokens_to_sample": 8,
  "temperature": 0.1,
  "prompt": "\n\nHuman: Hello, how are you?\n\nAssistant:"
}
Exampletext

text

{
  "type": "completion",
  "id": "982044c5-760c-4c8d-8936-f906b5cedc26",
  "completion": " Hey there! I'm doing great, thanks",
  "stop_reason": "max_tokens",
  "model": "grok-3"
}

最后更新:2026 年 9 月 2 日