Claude Messages API Anthropic Claude Messages 原生格式接口，调用 Claude 系列模型。
接口地址：POST /v1/messages
功能描述：使用 Anthropic Claude Messages 格式创建对话响应。
认证方式：Bearer Token
Authorization: Bearer YOUR_API_TOKEN
参数名 类型 必填 说明 示例 anthropic-version string 是 Anthropic API 版本 2023-06-01 Authorization string 是 Bearer Token Bearer sk-xxx
参数名 类型 必填 说明 示例 model string 是 模型名称 claude-opus-4-7 messages array 是 对话消息列表 - system string/array 否 系统提示词 - max_tokens integer 是 最大生成 token 数 1024 temperature number 否 温度 (0-1) 0.7 top_p number 否 Nucleus sampling 0.9 top_k integer 否 Top-k sampling 40 stream boolean 否 是否流式输出 true stop_sequences array 否 停止序列 - tools array 否 工具定义 - tool_choice object 否 工具选择策略 - thinking object 否 思考模式配置 - metadata object 否 元数据 -
参数名 类型 必填 说明 role string 是 角色，可选值：user, assistant content string/array 是 消息内容
参数名 类型 必填 说明 name string 是 工具名称 description string 否 工具描述 input_schema object 是 工具输入参数 schema
参数名 类型 必填 说明 type string 是 类型：enabled, disabled budget_tokens integer 否 思考预算 token 数
参数名 类型 说明 id string 消息 ID type string 类型，固定为 message role string 角色，固定为 assistant content array 内容列表 model string 模型名称 stop_reason string 停止原因 usage object Token 使用情况
参数名 类型 说明 type string 内容类型：text, tool_use text string 文本内容（type 为 text 时） id string 工具调用 ID（type 为 tool_use 时） name string 工具名称（type 为 tool_use 时） input object 工具输入（type 为 tool_use 时）
参数名 类型 说明 input_tokens integer 输入 tokens output_tokens integer 输出 tokens cache_creation_input_tokens integer 缓存创建 tokens cache_read_input_tokens integer 缓存读取 tokens
curl -X POST https://api.onedim.ai/v1/messages \ -H " Content-Type: application/json " \ -H " anthropic-version: 2023-06-01 " \ -H " Authorization: Bearer YOUR_API_TOKEN " \ -d ' { "model": "claude-opus-4-7", "max_tokens": 1024, "messages": [ {"role": "user", "content": "Hello, Claude!"} ] } ' 响应示例：
{ " id " : " msg_01X... " , " type " : " message " , " role " : " assistant " , " content " : [ { " type " : " text " , " text " : " Hello! How can I help you today? " } ], " model " : " claude-opus-4-7 " , " stop_reason " : " end_turn " , " usage " : { " input_tokens " : 10 , " output_tokens " : 12 , " cache_creation_input_tokens " : 0 , " cache_read_input_tokens " : 0 } } curl -X POST https://api.onedim.ai/v1/messages \ -H " Content-Type: application/json " \ -H " anthropic-version: 2023-06-01 " \ -H " Authorization: Bearer YOUR_API_TOKEN " \ -d ' { "model": "claude-opus-4-7", "max_tokens": 1024, "system": "You are a professional translator. Translate all user input to Chinese.", "messages": [ {"role": "user", "content": "Hello, how are you today?"} ] } ' import anthropic client = anthropic . Anthropic ( api_key = " YOUR_API_TOKEN " , base_url = " https://ezmodel.cloud " ) with client . messages . stream ( model = " claude-opus-4-7 " , max_tokens = 1024 , messages =[{ " role " : " user " , " content " : " 讲一个故事 " }] ) as stream : for text in stream . text_stream : print ( text , end = "" , flush = True ) curl -X POST https://api.onedim.ai/v1/messages \ -H " Content-Type: application/json " \ -H " anthropic-version: 2023-06-01 " \ -H " Authorization: Bearer YOUR_API_TOKEN " \ -d ' { "model": "claude-opus-4-7", "max_tokens": 1024, "tools": [ { "name": "get_weather", "description": "获取指定城市的天气信息", "input_schema": { "type": "object", "properties": { "city": { "type": "string", "description": "城市名称" } }, "required": ["city"] } } ], "messages": [ {"role": "user", "content": "北京今天天气怎么样？"} ] } ' import anthropic import base64 client = anthropic . Anthropic ( api_key = " YOUR_API_TOKEN " , base_url = " https://ezmodel.cloud " ) # 使用 URL message = client . messages . create ( model = " claude-opus-4-7 " , max_tokens = 1024 , messages =[ { " role " : " user " , " content " : [ { " type " : " image " , " source " : { " type " : " url " , " url " : " https://example.com/image.jpg " } }, { " type " : " text " , " text " : " 描述这张图片 " } ] } ] ) # 或使用 Base64 with open ( " image.jpg " , " rb " ) as f : image_data = base64 . standard_b64encode ( f . read ()). decode ( " utf-8 " ) message = client . messages . create ( model = " claude-opus-4-7 " , max_tokens = 1024 , messages =[ { " role " : " user " , " content " : [ { " type " : " image " , " source " : { " type " : " base64 " , " media_type " : " image/jpeg " , " data " : image_data } }, { " type " : " text " , " text " : " 描述这张图片 " } ] } ] ) 模型 上下文窗口 说明 claude-opus-4-7 200K / 1M 最强能力，适合复杂任务 claude-sonnet-4-6 200K / 1M 速度与能力平衡，日常首选
特性 Claude Messages OpenAI Chat 系统消息 独立 system 字段 messages 数组中 响应格式 content 数组 message.content 字符串 必填参数 max_tokens 必填 max_tokens 可选 停止原因 stop_reason finish_reason
如果你习惯使用 OpenAI 格式，也可以使用 /v1/chat/completions 接口调用 Claude 模型，系统会自动转换格式。