Claude Messages API

Inference Space provides Claude family conversational models, including the latest Opus, Sonnet, and Haiku generations. You can connect in either of two ways: the Anthropic-native Messages API or the OpenAI-compatible API. Both interfaces share the same account, API key, and billing system.

Overview

• Anthropic-native Messages: POST https://ai.inf.space/v1/messages, suitable for applications already using the Anthropic SDK or Claude ecosystem tools. You can reuse native capabilities such as messages, system, tools, thinking, and cache_control directly.
• OpenAI-compatible: POST https://ai.inf.space/v1/chat/completions, suitable for applications that already use an OpenAI-compatible client and want to switch to Claude without code changes.
• The model family includes Claude Opus, Sonnet, and Haiku: Opus targets complex programming and deep reasoning, Sonnet targets general-purpose tasks and everyday coding, and Haiku targets high concurrency and fast responses.

Authentication and Base URL

The API base URL is https://ai.inf.space; the browser console uses the same host. Both interfaces use the Authorization: Bearer header:

Native Messages requests must also include anthropic-version: 2023-06-01 and content-type: application/json. Create an API key in the console, store it in the AI_TOS_API_KEY environment variable, and then make the request.

Calling Messages

This is the Anthropic-native interface. max_tokens is required and limits the maximum number of output tokens in a single response.

curl https://ai.inf.space/v1/messages \
  -H "Authorization: Bearer $AI_TOS_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-x",
    "max_tokens": 1024,
    "system": "You are a concise Chinese assistant.",
    "messages": [
      { "role": "user", "content": "Introduce yourself in three sentences." }
    ]
  }'

import os, requests

resp = requests.post(
    "https://ai.inf.space/v1/messages",
    headers={
        "Authorization": f"Bearer {os.environ['AI_TOS_API_KEY']}",
        "anthropic-version": "2023-06-01",
        "content-type": "application/json",
    },
    json={
        "model": "claude-sonnet-4-x",
        "max_tokens": 1024,
        "system": "You are a concise Chinese assistant.",
        "messages": [
            {"role": "user", "content": "Introduce yourself in three sentences."}
        ],
    },
)
print(resp.json())

const resp = await fetch("https://ai.inf.space/v1/messages", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AI_TOS_API_KEY}`,
    "anthropic-version": "2023-06-01",
    "content-type": "application/json",
  },
  body: JSON.stringify({
    model: "claude-sonnet-4-x",
    max_tokens: 1024,
    system: "You are a concise Chinese assistant.",
    messages: [{ role: "user", content: "Introduce yourself in three sentences." }],
  }),
});
console.log(await resp.json());

OpenAI-Compatible Calls

If you already have an OpenAI-compatible client, switch the base URL, API key, and model ID to use Claude. In this interface, the system prompt is passed as a message with role: "system".

curl https://ai.inf.space/v1/chat/completions \
  -H "Authorization: Bearer $AI_TOS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-x",
    "messages": [
      { "role": "system", "content": "You are a concise Chinese assistant." },
      { "role": "user", "content": "Introduce yourself in three sentences." }
    ]
  }'

Parameters

The following parameters apply to the Anthropic-native Messages API.

The OpenAI-compatible API uses standard OpenAI fields such as messages, max_tokens, temperature, top_p, stop, stream, tools, and tool_choice.

Streaming

When stream: true is set, the response is returned in chunks as Server-Sent Events (SSE) for real-time rendering. The main event types are:

• message_start: Starts a message and carries the initial message metadata.
• content_block_delta: Content delta; text fragments are in delta.text.
• message_delta: Message-level delta carrying stop_reason and cumulative usage.
• message_stop: Ends the message.

curl https://ai.inf.space/v1/messages \
  -H "Authorization: Bearer $AI_TOS_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-x",
    "max_tokens": 1024,
    "stream": true,
    "messages": [
      { "role": "user", "content": "Write a short poem about clouds." }
    ]
  }'

Tool Calls

Declare callable functions with tools. When needed, the model returns a tool_use content block. After the application executes the tool, add the result to the next messages turn as a tool_result block so the model can produce the final answer.

curl https://ai.inf.space/v1/messages \
  -H "Authorization: Bearer $AI_TOS_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-x",
    "max_tokens": 1024,
    "tools": [
      {
        "name": "get_weather",
        "description": "Get the weather for a specified city",
        "input_schema": {
          "type": "object",
          "properties": {
            "city": { "type": "string", "description": "City name" }
          },
          "required": ["city"]
        }
      }
    ],
    "messages": [
      { "role": "user", "content": "What's the weather like in Hefei today?" }
    ]
  }'

The returned tool_use block contains id, name, and input. After executing the tool, add the result and send another request using this structure:

{
  "role": "user",
  "content": [
    {
      "type": "tool_result",
      "tool_use_id": "toolu_xxx",
      "content": "Sunny, 26°C"
    }
  ]
}

Prompt Caching

For stable, repeatedly used content such as system prompts and long documents, add cache_control to a content block to use prompt caching, reduce input costs, and improve response speed.

{
  "system": [
    {
      "type": "text",
      "text": "(A large block of fixed business context and rules...)",
      "cache_control": { "type": "ephemeral" }
    }
  ]
}

Writing to the cache produces cache_creation_input_tokens. Later cache hits are counted in cache_read_input_tokens, and cached input is billed at a lower price.

Responses and Usage

Key fields in a native Messages response:

{
  "id": "msg_xxx",
  "type": "message",
  "role": "assistant",
  "content": [
    { "type": "text", "text": "Hello, I am a Chinese assistant..." }
  ],
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 24,
    "output_tokens": 38,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0
  }
}

Billing

The Claude family follows Inference Space LLM billing rules and is metered in tokens, in units of 1M tokens:

• input_tokens (non-cached input), cached_input_tokens (cached input), and output_tokens are billed separately.
• Cached input is billed at a lower price. When an explicit cache price exists, that price is used; otherwise, the lower discount for the same context tier applies.
• Context tiers are determined by the input tokens in the request: ≤128k, ≤256k, and >256k. Each tier has different unit pricing.
• Reasoning (extended thinking), tool results, and retrieved context are ultimately reflected in input or output tokens.

Integration Recommendations

• Keep API keys on the server in production; do not expose them in browsers or client applications.
• Use a different key for each application to simplify auditing, quota management, and revocation.
• Estimate tokens from representative business samples before setting quotas and alerts; include tool calls, retrieved context, and thinking tokens in the budget.
• Manage the model ID as a configuration value so you can smoothly switch to new versions as the console is updated.