API

Chat Completions

The OpenAI-compatible /v1/chat/completions endpoint.

POST /v1/chat/completions is a drop-in implementation of the OpenAI Chat Completions API. Any OpenAI SDK or HTTP client works unchanged after a base URL swap.

Request

bash

curl https://api.tokenrouter.io/v1/chat/completions \
  -H "Authorization: Bearer tr_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5-mini",
    "messages": [
      {"role": "system", "content": "You are a concise assistant."},
      {"role": "user", "content": "What is an AI gateway?"}
    ],
    "temperature": 0.7,
    "max_tokens": 200
  }'

Response

json

{
  "id": "chatcmpl-9f3a1b2c",
  "object": "chat.completion",
  "created": 1754400000,
  "model": "openai/gpt-5-mini",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "An AI gateway is a single API layer that sits between your apps and model providers..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 27,
    "completion_tokens": 42,
    "total_tokens": 69
  }
}

Streaming

Set "stream": true to receive Server-Sent Events. Each event is a chat.completion.chunk; the stream terminates with a literal data: [DONE] line.

bash

curl -N https://api.tokenrouter.io/v1/chat/completions \
  -H "Authorization: Bearer tr_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4-5",
    "stream": true,
    "stream_options": {"include_usage": true},
    "messages": [{"role": "user", "content": "Write a haiku about budgets."}]
  }'

text

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hard"}}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" caps"}}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","choices":[],"usage":{"prompt_tokens":14,"completion_tokens":19,"total_tokens":33}}

data: [DONE]

With stream_options.include_usage set, the final chunk before [DONE] carries the full usage object and an empty choices array — exactly like OpenAI.

Tool calling

json

{
  "model": "openai/gpt-5-mini",
  "messages": [{"role": "user", "content": "What is the weather in Berlin?"}],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get current weather for a city",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {"type": "string"}
          },
          "required": ["city"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}

Tool calls come back in choices[0].message.tool_calls; return results as role: "tool" messages. Tool calling is translated natively for every provider that supports it.

Reasoning models from the DeepSeek family return their chain-of-thought in message.reasoning_content (and delta.reasoning_content while streaming), matching DeepSeek’s API shape.