kimi-k3

curl --request POST \
  --url https://www.anyfast.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "kimi-k3",
  "messages": [
    {
      "role": "user",
      "content": "Explain Kimi K3 in one sentence."
    }
  ],
  "reasoning_effort": "max"
}
'

{
  "id": "cmpl-kimi-k3-example",
  "object": "chat.completion",
  "created": 1784736000,
  "model": "moonshotai/kimi-k3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "reasoning": "The user requested a concise introduction.",
        "reasoning_details": [
          {
            "type": "reasoning.text",
            "text": "The user requested a concise introduction.",
            "format": "unknown",
            "index": 0
          }
        ],
        "content": "Kimi K3 is Moonshot AI's flagship multimodal reasoning model with a 1M-token context window."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 18,
    "completion_tokens": 32,
    "total_tokens": 50,
    "prompt_tokens_details": {
      "cached_tokens": 0
    }
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

POST

chat

completions

curl --request POST \
  --url https://www.anyfast.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "kimi-k3",
  "messages": [
    {
      "role": "user",
      "content": "Explain Kimi K3 in one sentence."
    }
  ],
  "reasoning_effort": "max"
}
'

{
  "id": "cmpl-kimi-k3-example",
  "object": "chat.completion",
  "created": 1784736000,
  "model": "moonshotai/kimi-k3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "reasoning": "The user requested a concise introduction.",
        "reasoning_details": [
          {
            "type": "reasoning.text",
            "text": "The user requested a concise introduction.",
            "format": "unknown",
            "index": 0
          }
        ],
        "content": "Kimi K3 is Moonshot AI's flagship multimodal reasoning model with a 1M-token context window."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 18,
    "completion_tokens": 32,
    "total_tokens": 50,
    "prompt_tokens_details": {
      "cached_tokens": 0
    }
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

{
  "error": {
    "type": "<string>",
    "message": "<string>",
    "param": "<string>",
    "code": "<string>"
  }
}

Kimi K3 支持 100 万 token 上下文、文本/图片输入、可配置强度的始终开启推理、流式输出、结构化输出、Partial Mode 和函数工具。

使用顶层 reasoning_effort；K2.x 的 thinking 对象不会生效。采样参数为固定值，请省略 temperature、top_p、n、presence_penalty 和 frequency_penalty。

授权

Authorization

string

header

必填

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

请求体

application/json

model

enum<string>

默认值:kimi-k3

必填

模型 ID，固定为 kimi-k3。

可用选项:

kimi-k3

messages

object[]

必填

对话历史。后续轮次必须完整传回 assistant 消息，包括 reasoning、reasoning_details 和 tool_calls。

Minimum array length: 1

Show child attributes

reasoning_effort

enum<string>

默认值:max

控制推理强度。K3 始终开启思考与 Preserved Thinking，无法关闭思考。

可用选项:

low,

high,

max

max_completion_tokens

integer

默认值:131072

最大输出 Token 数。输入与输出之和不能超过 100 万 Token 上下文窗口。 max_tokens 已弃用。

必填范围: 1 <= x <= 1048576

response_format

object

Show child attributes

stop

最多 5 个停止字符串，每个字符串不超过 32 字节。停止词可能在最终内容生成前终止推理。

stream

boolean

默认值:false

是否通过 Server-Sent Events 流式返回增量内容。

stream_options

object

流式响应选项。

Show child attributes

tools

object[]

可供模型调用的函数工具列表。

Show child attributes

tool_choice

enum<string>

默认值:auto

让模型自行决定、禁用工具调用，或要求至少调用一个工具。

可用选项:

auto,

none,

required

prompt_cache_key

string

稳定的会话或任务标识，用于提高前缀缓存命中率。

safety_identifier

string

最终用户的稳定标识。建议使用哈希值，不要传入可直接识别用户的信息。

响应

成功生成响应。stream 为 true 时通过 SSE 返回 chat.completion.chunk，并以 data: [DONE] 结束。

string

必填

object

enum<string>

必填

可用选项:

chat.completion

created

integer

必填

Unix 时间戳。

model

string

必填

响应中可能返回上游规范模型 ID。

示例:

"moonshotai/kimi-k3"

choices

object[]

必填

Show child attributes

provider

string

示例:

"Moonshot AI"

system_fingerprint

string | null

service_tier

string | null

usage

object

Show child attributes

minimax-m2.1 kimi-k2.6

概览

原生兼容端点

模型 API

授权

请求体

响应