Claude API Messages endpoint - Create Claude 3.5 Sonnet, Claude 3 Opus conversations through UniterRouter. Anthropic-compatible API with streaming and tool use support.

Messages API

Create a message using Claude models. This endpoint provides native Anthropic API compatibility for applications that prefer the Anthropic SDK.

Alternative: You can also access Claude models through the OpenAI-compatible API, which is recommended for most use cases.

Endpoint

POST https://api.uniterouter.ai/v1/messages

Authentication

UniterRouter supports two authentication methods:

Option 1: Anthropic Native (x-api-key)

x-api-key: your_uniterouter_api_key
anthropic-version: 2023-06-01

Option 2: Bearer Token

Authorization: Bearer your_uniterouter_api_key
anthropic-version: 2023-06-01

Both authentication methods work identically. Use x-api-key for Anthropic SDK compatibility, or Bearer for tools like Claude Code.

Request Body

UniterRouter 完全兼容 Anthropic 官方 API，支持所有参数。

基础参数

Parameter	Type	Required	Description
`model`	string	Yes	Model ID (e.g., `claude-sonnet-4-20250514`)
`messages`	array	Yes	Array of message objects
`max_tokens`	integer	Yes	Maximum tokens to generate

系统提示

Parameter	Type	Required	Description
`system`	string \| array	No	System prompt. Can be a string or array of `TextBlockParam`

采样参数

Parameter	Type	Required	Description
`temperature`	number	No	Sampling temperature (0.0-1.0, default: 1.0)
`top_p`	number	No	Nucleus sampling parameter
`top_k`	integer	No	Top-k sampling parameter
`stop_sequences`	array	No	Custom stop sequences

流式与元数据

Parameter	Type	Required	Description
`stream`	boolean	No	Enable streaming (default: false)
`metadata`	object	No	Request metadata with optional `user_id`
`service_tier`	string	No	Service tier: `auto` or `standard_only`

工具调用

Parameter	Type	Required	Description
`tools`	array	No	Tool definitions (custom, bash, text_editor, web_search)
`tool_choice`	object	No	Tool selection: `auto`, `any`, `tool`, or `none`

扩展思考 (Extended Thinking)

Parameter	Type	Required	Description
`thinking`	object	No	Extended thinking configuration

Thinking Configuration:

{
  "thinking": {
    "type": "enabled",
    "budget_tokens": 10000
  }
}

Extended thinking allows Claude to show its reasoning process. Requires budget_tokens ≥ 1024 and less than max_tokens.

Message Format

{
  "messages": [
    {
      "role": "user",
      "content": "Hello, Claude!"
    }
  ]
}

Message with Images

{
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "image",
          "source": {
            "type": "base64",
            "media_type": "image/jpeg",
            "data": "/9j/4AAQSkZJRg..."
          }
        },
        {
          "type": "text",
          "text": "What's in this image?"
        }
      ]
    }
  ]
}

Example Request

curl https://api.uniterouter.ai/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: $UNITEROUTER_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "max_tokens": 1024,
    "system": "You are a helpful assistant.",
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ]
  }'

import anthropic

client = anthropic.Anthropic(
    api_key="your_uniterouter_api_key",
    base_url="https://api.uniterouter.ai"
)

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Hello, Claude!"}
    ]
)

print(message.content[0].text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: process.env.UNITEROUTER_API_KEY,
  baseURL: "https://api.uniterouter.ai"
});

const message = await client.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  system: "You are a helpful assistant.",
  messages: [
    { role: "user", content: "Hello, Claude!" }
  ]
});

console.log(message.content[0].text);

Response

{
  "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! How can I help you today?"
    }
  ],
  "model": "claude-3-5-sonnet-20241022",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 25,
    "output_tokens": 15
  }
}

Streaming

Enable streaming for real-time responses:

with client.messages.stream(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Tell me a story"}]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Stream Events

Event Type	Description
`message_start`	Message creation started
`content_block_start`	Content block started
`content_block_delta`	Content chunk received
`content_block_stop`	Content block finished
`message_delta`	Message metadata update
`message_stop`	Message completed

Tool Use (Function Calling)

Claude supports tool use for structured interactions:

tools = [
    {
        "name": "get_weather",
        "description": "Get the current weather for a location",
        "input_schema": {
            "type": "object",
            "properties": {
                "location": {
                    "type": "string",
                    "description": "City name"
                }
            },
            "required": ["location"]
        }
    }
]

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    tools=tools,
    messages=[
        {"role": "user", "content": "What's the weather in Paris?"}
    ]
)

# Check if tool was called
for block in message.content:
    if block.type == "tool_use":
        print(f"Tool: {block.name}")
        print(f"Input: {block.input}")

Vision

Send images to Claude for analysis:

import base64

with open("image.jpg", "rb") as f:
    image_data = base64.standard_b64encode(f.read()).decode("utf-8")

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "base64",
                        "media_type": "image/jpeg",
                        "data": image_data
                    }
                },
                {
                    "type": "text",
                    "text": "Describe this image in detail."
                }
            ]
        }
    ]
)

Available Models

Model	Description	Context	Max Output
`claude-sonnet-4-20250514`	Latest Claude 4 Sonnet	200K	64K
`claude-opus-4-20250514`	Most powerful Claude 4	200K	32K
`claude-3-7-sonnet-20250219`	Claude 3.7 with extended thinking	200K	128K
`claude-3-5-sonnet-20241022`	Claude 3.5 Sonnet	200K	8K
`claude-3-5-haiku-20241022`	Fast and affordable	200K	8K
`claude-3-opus-20240229`	Claude 3 Opus	200K	4K
`claude-3-haiku-20240307`	Previous Haiku	200K	4K

Error Codes

Code	Description
`400`	Invalid request parameters
`401`	Invalid API key
`403`	Permission denied
`404`	Model not found
`429`	Rate limit exceeded
`500`	Internal server error
`529`	API overloaded

Chat Completions (OpenAI) - OpenAI-compatible endpoint
Vision - Image input guide
Reasoning - Extended thinking

Messages

On this page