Skip to content

Kimi 模型接口文档 ​

兼容 OpenAI Chat Completions 接口,支持 Kimi(Moonshot AI)系列模型的普通消息、多模态输入与流式输出。

Base URL:https://api.agentpivot.net/v1/


登录平台 ​


接口概览 ​

方法路径说明
POST/chat/completionsKimi 对话接口(Chat Completions)
POST/responsesKimi 响应式接口(Responses,OpenAI 兼容)

鉴权 ​

使用 API Key:

Header说明
x-api-key: <API_KEY>使用 API Key

示例:

bash
curl -X POST "https://api.agentpivot.net/v1/chat/completions" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k2.6",
    "messages": [{
      "role": "user",
      "content": "你好"
    }]
  }'

支持模型 ​

模型名称说明
kimi-k3旗舰模型,2.8 万亿参数,1M token 上下文,原生视觉理解,面向长程编程与端到端知识工作
kimi-k2.7-code-highspeed面向代码场景的 Coding 模型,256K 上下文,文本/图片/视频输入,思考模式,更高输出速度
kimi-k2.6通用能力强,256K 上下文,文本/图片/视频输入,思考与非思考模式,适合通用对话、Agent、视觉理解与复杂推理

POST /chat/completions ​

请求 ​

  • Content-Type:application/json

Body ​

字段类型必填说明
modelstring是模型名称
messagesarray是对话消息列表
streamboolean否是否启用流式输出
max_tokensnumber否最大输出 Token 数
temperaturenumber否采样温度
top_pnumber否Top-P 采样
reasoning_effortstring否推理强度,kimi-k3 支持 low / high / max(默认 max)

messages ​

字段类型说明
rolestringsystem | user | assistant | tool
contentstring | array消息内容;支持纯文本字符串,或用于多模态输入的对象数组

content 多模态输入 ​

content 支持对象数组形式,数组元素通过 type 区分:

type说明
text文本内容
image_url图片(base64 data URL 或文件引用)
video_url视频(base64 data URL 或文件引用,kimi-k3 / kimi-k2.6 / kimi-k2.7-code 支持)

请求示例 ​

1. kimi-k3(旗舰,配置推理强度) ​

bash
#!/usr/bin/env bash

API_KEY="YOUR_API_KEY"

curl -s https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: ${API_KEY}" \
  -d '{
    "model": "kimi-k3",
    "messages": [
      {
        "role": "user",
        "content": "分析未来五年 AI 发展趋势"
      }
    ],
    "reasoning_effort": "high",
    "stream": false
  }' | jq

2. kimi-k2.6(通用对话) ​

bash
#!/usr/bin/env bash

API_KEY="YOUR_API_KEY"

curl -s https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: ${API_KEY}" \
  -d '{
    "model": "kimi-k2.6",
    "messages": [
      {
        "role": "user",
        "content": "你好"
      }
    ],
    "stream": false
  }' | jq

3. kimi-k2.7-code-highspeed(代码生成) ​

bash
#!/usr/bin/env bash

API_KEY="YOUR_API_KEY"

curl -s https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: ${API_KEY}" \
  -d '{
    "model": "kimi-k2.7-code-highspeed",
    "messages": [
      {
        "role": "user",
        "content": "写一个快速排序的 Python 实现"
      }
    ],
    "stream": false
  }' | jq

4. 多模态输入(图片理解) ​

bash
#!/usr/bin/env bash

API_KEY="YOUR_API_KEY"

curl -s https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: ${API_KEY}" \
  -d '{
    "model": "kimi-k2.6",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "image_url",
            "image_url": {
              "url": "data:image/jpeg;base64,/9j/4AAQSkZJRg..."
            }
          },
          {
            "type": "text",
            "text": "请描述这张图片"
          }
        ]
      }
    ],
    "stream": false
  }' | jq

流式输出 ​

kimi-k3 ​

bash
curl -N https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "model": "kimi-k3",
    "messages": [
      {
        "role": "user",
        "content": "详细解释 Transformer 工作原理"
      }
    ],
    "stream": true
  }'

kimi-k2.6 ​

bash
curl -N https://api.agentpivot.net/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "model": "kimi-k2.6",
    "messages": [
      {
        "role": "user",
        "content": "写一个快速排序"
      }
    ],
    "stream": true
  }'

响应示例 ​

json
{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "kimi-k2.6",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "你好!我是 Kimi,有什么可以帮你的吗?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 15,
    "total_tokens": 27
  }
}

POST /responses(Responses 接口) ​

Kimi 同样兼容 OpenAI 的 Responses 接口,面向多轮对话、工具调用与 Agent 场景。与 Chat Completions 相比,请求/响应结构与字段命名不同。

路径:POST /responses

与 /chat/completions 的关键差异 ​

Chat CompletionsResponses
messagesinput(字符串或输入条目列表)
system 消息放入 messagesinstructions(作为第一条 system 消息)
max_tokensmax_output_tokens
响应 choices[].message.content响应 output[](含 message / function_call 等条目)

Body(顶层参数) ​

字段类型必填说明
modelstring是模型名称
inputstring | array是*输入内容(字符串或输入条目列表),input 与 instructions 至少填一个
instructionsstring否系统指令,作为第一条 system 消息
streamboolean否是否启用流式输出(SSE)
temperaturenumber否采样温度
top_pnumber否Top-P 采样
max_output_tokensnumber否最大输出 Token 数
reasoning_effortstring否推理强度(kimi-k3 支持 low / high / max)

*input 与 instructions 至少提供一个。

请求示例 ​

bash
curl -s https://api.agentpivot.net/v1/responses \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "model": "kimi-k3",
    "instructions": "你是一个严谨的技术顾问。",
    "input": "什么是响应式接口?",
    "reasoning_effort": "high"
  }' | jq

流式输出(SSE) ​

bash
curl -N https://api.agentpivot.net/v1/responses \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "model": "kimi-k2.6",
    "instructions": "You are a helpful assistant.",
    "input": "写一个快速排序",
    "stream": true
  }'

流式返回为 SSE 事件序列,以 response.completed / response.incomplete / response.failed 结束(没有 data: [DONE])。关键事件:response.created、response.output_text.delta、response.output_text.done、response.completed。

响应示例 ​

json
{
  "id": "resp_xxx",
  "object": "response",
  "created_at": 1700000000,
  "status": "completed",
  "model": "kimi-k3",
  "output": [
    {
      "type": "message",
      "id": "msg_xxx",
      "status": "completed",
      "role": "assistant",
      "content": [
        {
          "type": "output_text",
          "text": "响应式接口(Responses API)是 OpenAI 兼容的新一代接口。",
          "annotations": []
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 20,
    "output_tokens": 25,
    "total_tokens": 45,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 8
    }
  }
}

测试脚本 ​

bash
export API_KEY="YOUR_API_KEY"

curl -s https://api.agentpivot.net/v1/chat/completions \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model":"kimi-k2.6",
    "messages":[
      {
        "role":"user",
        "content":"介绍一下 Kimi"
      }
    ]
  }' | jq