messages

Create message (Claude native API)

Create a message using Anthropic Claude’s native API format, supports streaming

Request

POST /messages

Request Body

Field	Type	Required	Description
model	string	Yes	Claude model name
messages	Array<AnthropicMessageMessage>	Yes	Messages list, first message must be a user message
system	string	No	System prompt
max_tokens	integer	No	Maximum number of output tokens (optional). If not provided, the relay service or API may use a default value. Different models have different maximum values.
stop_sequences	Array	No	Stop sequences
stream	boolean	No	-
temperature	number	No	-
top_p	number	No	-
top_k	integer	No	Top-k sampling parameter. Only sample from the top K options for each subsequent token.
tools	Array<AnthropicTool>	No	-
tool_choice	string (none, auto, any)	No	-
metadata	object	No	An object describing metadata about the request. Can be used for tracking, identification, or filtering purposes. Common use cases: user_id, session_id, request_id, etc.
thinking	object	No	Configuration for extended thinking (Claude 3.7+). When enabled, the model will spend more time thinking before responding.
service_tier	string (auto, standard_only)	No	Service tier for request processing: - auto: Automatically select between standard and priority capacity - standard_only: Only use standard capacity (may have longer wait times during high load)

Request Examples

Simple Claude message

{
  "model": "claude-opus-4.5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Hello, Claude!"
        }
      ]
    }
  ]
}

Message with system prompt

{
  "model": "claude-sonnet-4.5",
  "max_tokens": 2048,
  "system": "You are a helpful coding assistant.",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "How do I write a Python function?"
        }
      ]
    }
  ],
  "temperature": 0.7
}

Image understanding (Vision)

{
  "model": "claude-sonnet-4.5",
  "max_tokens": 2048,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What do you see in this image?"
        },
        {
          "type": "image",
          "source": {
            "type": "base64",
            "media_type": "image/jpeg",
            "data": "/9j/4AAQSkZJRg..."
          }
        }
      ]
    }
  ],
  "temperature": 0.5
}

With tool calls

{
  "model": "claude-opus-4.5",
  "max_tokens": 4096,
  "system": "You are a helpful assistant with access to tools.",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What's the weather in Paris and what time is it there?"
        }
      ]
    }
  ],
  "tools": [
    {
      "name": "get_weather",
      "description": "Get current weather information for a location",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string",
            "description": "City name"
          },
          "units": {
            "type": "string",
            "enum": [
              "celsius",
              "fahrenheit"
            ],
            "description": "Temperature units"
          }
        },
        "required": [
          "location"
        ]
      }
    },
    {
      "name": "get_time",
      "description": "Get current time for a location",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string",
            "description": "City or timezone"
          }
        },
        "required": [
          "location"
        ]
      }
    }
  ],
  "tool_choice": "auto"
}

Tool use result response

{
  "model": "claude-sonnet-4.5",
  "max_tokens": 2048,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What's the weather in Tokyo?"
        }
      ]
    },
    {
      "role": "assistant",
      "content": [
        {
          "type": "tool_use",
          "id": "toolu_01A",
          "name": "get_weather",
          "input": {
            "location": "Tokyo",
            "units": "celsius"
          }
        }
      ]
    },
    {
      "role": "user",
      "content": [
        {
          "type": "tool_result",
          "tool_use_id": "toolu_01A",
          "content": "{\"temperature\": 22, \"condition\": \"cloudy\", \"humidity\": 70}"
        }
      ]
    }
  ],
  "tools": [
    {
      "name": "get_weather",
      "description": "Get weather information",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string"
          },
          "units": {
            "type": "string"
          }
        },
        "required": [
          "location"
        ]
      }
    }
  ]
}

Multi-turn conversation with tool calls

{
  "model": "claude-opus-4.5",
  "max_tokens": 4096,
  "system": "You are a research assistant. Use tools when needed to provide accurate information.",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Find me information about the latest SpaceX launch"
        }
      ]
    },
    {
      "role": "assistant",
      "content": [
        {
          "type": "tool_use",
          "id": "toolu_01B",
          "name": "web_search",
          "input": {
            "query": "latest SpaceX launch",
            "max_results": 5
          }
        }
      ]
    },
    {
      "role": "user",
      "content": [
        {
          "type": "tool_result",
          "tool_use_id": "toolu_01B",
          "content": "{\"results\": [{\"title\": \"SpaceX Starship Launch\", \"date\": \"2024-12-15\", \"summary\": \"...\"}]}"
        }
      ]
    },
    {
      "role": "assistant",
      "content": [
        {
          "type": "text",
          "text": "Based on the search results, the latest SpaceX launch was..."
        }
      ]
    },
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What was the weather during that launch?"
        }
      ]
    }
  ],
  "tools": [
    {
      "name": "web_search",
      "description": "Search the web",
      "input_schema": {
        "type": "object",
        "properties": {
          "query": {
            "type": "string"
          },
          "max_results": {
            "type": "integer"
          }
        },
        "required": [
          "query"
        ]
      }
    },
    {
      "name": "get_weather",
      "description": "Get weather information",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string"
          },
          "date": {
            "type": "string"
          }
        },
        "required": [
          "location"
        ]
      }
    }
  ],
  "temperature": 0.7
}

Streaming

{
  "model": "claude-sonnet-4.5",
  "max_tokens": 3000,
  "stream": true,
  "system": "You are a creative writing assistant.",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Write a short story about a robot learning to paint"
        }
      ]
    }
  ],
  "temperature": 0.9
}

Specify specific tool

{
  "model": "claude-opus-4.5",
  "max_tokens": 2048,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Tell me about the weather"
        }
      ]
    }
  ],
  "tools": [
    {
      "name": "get_weather",
      "description": "Get weather information",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string"
          }
        },
        "required": [
          "location"
        ]
      }
    },
    {
      "name": "get_forecast",
      "description": "Get weather forecast",
      "input_schema": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string"
          },
          "days": {
            "type": "integer"
          }
        },
        "required": [
          "location"
        ]
      }
    }
  ],
  "tool_choice": {
    "type": "tool",
    "name": "get_weather"
  }
}

Complex multimodal conversation

{
  "model": "claude-opus-4.5",
  "max_tokens": 4096,
  "system": "You are an AI assistant helping with data analysis and visualization.",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Analyze this chart and tell me the trends"
        },
        {
          "type": "image",
          "source": {
            "type": "base64",
            "media_type": "image/png",
            "data": "iVBORw0KGgoAAAANS..."
          }
        }
      ]
    },
    {
      "role": "assistant",
      "content": [
        {
          "type": "text",
          "text": "I can see the chart shows a clear upward trend..."
        }
      ]
    },
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Can you fetch the latest data to compare?"
        }
      ]
    }
  ],
  "tools": [
    {
      "name": "fetch_data",
      "description": "Fetch latest dataset",
      "input_schema": {
        "type": "object",
        "properties": {
          "dataset_id": {
            "type": "string"
          },
          "date_range": {
            "type": "string"
          }
        },
        "required": [
          "dataset_id"
        ]
      }
    }
  ],
  "temperature": 0.5,
  "top_p": 0.9,
  "top_k": 40
}

Using stop sequences

{
  "model": "claude-sonnet-4.5",
  "max_tokens": 1500,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Generate a numbered list of programming tips"
        }
      ]
    }
  ],
  "stop_sequences": [
    "\n\n",
    "Conclusion"
  ],
  "temperature": 0.8
}

Successful response

Response Schema

Field	Type	Required	Description
id	string	Yes	-
type	string	Yes	-
role	string	Yes	-
content	Array<AnthropicContent>	Yes	-
model	string	Yes	-
stop_reason	object	Yes	Reason why the model stopped: - end_turn: Natural completion - max_tokens: Hit max_tokens limit - stop_sequence: Hit a stop sequence - tool_use: Model wants to use a tool - pause_turn: Long-running task paused (extended thinking) - refusal: Content policy violation
stop_sequence	object	No	-
usage	object	Yes	-

Response Example

{
  "id": "msg_123",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! It's nice to meet you. How can I help you today?"
    }
  ],
  "model": "claude-opus-4.5",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 12,
    "output_tokens": 15
  }
}

Code Examples

JavaScript (Fetch)

const response = await fetch('https://api.r9s.ai/v1/messages', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
  "model": "claude-opus-4.5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Hello, Claude!"
        }
      ]
    }
  ]
})
});

const data = await response.json();
console.log(data);

Python (requests)

import requests

url = "https://api.r9s.ai/v1/messages"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

response = requests.post(url, json={
  "model": "claude-opus-4.5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Hello, Claude!"
        }
      ]
    }
  ]
}, headers=headers)
data = response.json()
print(data)

cURL

curl -X POST "https://api.r9s.ai/v1/messages" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-opus-4.5","max_tokens":1024,"messages":[{"role":"user","content":[{"type":"text","text":"Hello, Claude!"}]}]}'

Schema Reference

AnthropicMessageRequest

Field	Type	Required	Description
model	string	Yes	Claude model name
messages	Array<AnthropicMessageMessage>	Yes	Messages list, first message must be a user message
system	string	No	System prompt
max_tokens	integer	No	Maximum number of output tokens (optional). If not provided, the relay service or API may use a default value. Different models have different maximum values.
stop_sequences	Array	No	Stop sequences
stream	boolean	No	-
temperature	number	No	-
top_p	number	No	-
top_k	integer	No	Top-k sampling parameter. Only sample from the top K options for each subsequent token.
tools	Array<AnthropicTool>	No	-
tool_choice	string (none, auto, any)	No	-
metadata	object	No	An object describing metadata about the request. Can be used for tracking, identification, or filtering purposes. Common use cases: user_id, session_id, request_id, etc.
thinking	object	No	Configuration for extended thinking (Claude 3.7+). When enabled, the model will spend more time thinking before responding.
service_tier	string (auto, standard_only)	No	Service tier for request processing: - auto: Automatically select between standard and priority capacity - standard_only: Only use standard capacity (may have longer wait times during high load)

AnthropicMessageMessage

Field	Type	Required	Description
role	string (user, assistant)	Yes	Role of the message sender (user or assistant)
content	Array<AnthropicContent>	Yes	Message content - can be array of content blocks or string

AnthropicContent

AnthropicTextContent

Field	Type	Required	Description
type	string	Yes	-
text	string	Yes	-

AnthropicImageContent

Field	Type	Required	Description
type	string	Yes	-
source	object	Yes	-

AnthropicImageSource

AnthropicImageSourceBase64

Field	Type	Required	Description
type	string	Yes	-
media_type	string (image/jpeg, image/png, image/gif, image/webp)	Yes	MIME type of the image
data	string	Yes	Base64-encoded image data

AnthropicImageSourceUrl

Field	Type	Required	Description
type	string	Yes	-
url	string	Yes	URL of the image to fetch

AnthropicDocumentContent

Field	Type	Required	Description
type	string	Yes	Document content type (PDFs, text files, etc.)
source	object	Yes	-

AnthropicDocumentSource

AnthropicDocumentSourceBase64

Field	Type	Required	Description
type	string	Yes	-
media_type	string (application/pdf, text/plain)	Yes	Document MIME type
data	string	Yes	Base64-encoded document data

AnthropicDocumentSourceUrl

Field	Type	Required	Description
type	string	Yes	-
url	string	Yes	URL of the document to fetch

AnthropicThinkingContent

Field	Type	Required	Description
type	string	Yes	Extended thinking output content
thinking	string	Yes	The model’s thinking process/reasoning
signature	string	No	Cryptographic signature for thinking content verification

AnthropicToolContent

AnthropicToolUseContent

Field	Type	Required	Description
type	string	Yes	-
id	string	Yes	Unique identifier for this tool use
name	string	Yes	Name of the tool being called
input	object	Yes	Input parameters for the tool

AnthropicToolResultContent

Field	Type	Required	Description
type	string	Yes	-
tool_use_id	string	Yes	ID of the tool use this result corresponds to
content	string	Yes	Result of the tool execution

AnthropicTool

Field	Type	Required	Description
name	string	Yes	-
description	string	No	-
input_schema	object	Yes	-

AnthropicInputSchema

Field	Type	Required	Description
type	string	Yes	-
properties	object	No	-
required	Array	No	-

AnthropicMessageResponse

Field	Type	Required	Description
id	string	Yes	-
type	string	Yes	-
role	string	Yes	-
content	Array<AnthropicContent>	Yes	-
model	string	Yes	-
stop_reason	object	Yes	Reason why the model stopped: - end_turn: Natural completion - max_tokens: Hit max_tokens limit - stop_sequence: Hit a stop sequence - tool_use: Model wants to use a tool - pause_turn: Long-running task paused (extended thinking) - refusal: Content policy violation
stop_sequence	object	No	-
usage	object	Yes	-

API Overview - Learn about authentication and basic information
models - View models related APIs
chat - View chat related APIs
responses - View responses related APIs
completions - View completions related APIs
edits - View edits related APIs
images - View images related APIs
embeddings - View embeddings related APIs
engine-embeddings - View engine-embeddings related APIs
moderations - View moderations related APIs
audio - View audio related APIs
search - View search related APIs
proxy - View proxy related APIs

messages

Create message (Claude native API)

Request

Request Body

Request Examples

Simple Claude message

Message with system prompt

Image understanding (Vision)

With tool calls

Tool use result response

Multi-turn conversation with tool calls

Streaming

Specify specific tool

Complex multimodal conversation

Using stop sequences

Response Schema

Response Example

Code Examples

JavaScript (Fetch)

Python (requests)

cURL

Schema Reference

AnthropicMessageRequest

AnthropicMessageMessage

AnthropicContent

AnthropicTextContent

AnthropicImageContent

AnthropicImageSource

AnthropicImageSourceBase64

AnthropicImageSourceUrl

AnthropicDocumentContent

AnthropicDocumentSource

AnthropicDocumentSourceBase64

AnthropicDocumentSourceUrl

AnthropicThinkingContent

AnthropicToolContent

AnthropicToolUseContent

AnthropicToolResultContent

AnthropicTool

AnthropicInputSchema

AnthropicMessageResponse

Related APIs