Anthropic Messages

Anthropic Messages API endpoint (Claude API compatible).

This endpoint accepts Anthropic-style requests and transforms them to work
with OpenAI-compatible providers (OpenRouter, Portkey, Featherless).

Key differences from OpenAI Chat Completions:

Uses 'messages' array but 'system' is a separate parameter

'max_tokens' is REQUIRED (not optional like in OpenAI)

Returns Anthropic-style response with 'content' array and 'stop_reason'

Supports 'stop_sequences' instead of 'stop'

Supports 'top_k' parameter (Anthropic-specific, logged but not used)

Example request:

{
  "model": "claude-sonnet-4-5-20250929",
  "max_tokens": 1024,
  "messages": [
    {"role": "user", "content": "Hello, Claude!"}
  ]
}

Example response:

{
  "id": "msg-123",
  "type": "message",
  "role": "assistant",
  "content": [{"type": "text", "text": "Hello! How can I help?"}],
  "model": "claude-sonnet-4-5-20250929",
  "stop_reason": "end_turn",
  "usage": {"input_tokens": 10, "output_tokens": 12}
}

Bearer Token

Provide your bearer token in the

Authorization

header when making requests to protected resources.

Example:

Authorization: Bearer ********************

curl --location --request POST '/v1/messages?session_id' \ --header 'Authorization: Bearer <token>' \ --header 'Content-Type: application/json' \ --data-raw '{ "model": "string", "messages": [ { "role": "string", "content": "string" } ], "max_tokens": 0, "system": "string", "temperature": 0, "top_p": 0, "top_k": 0, "stop_sequences": [ "string" ], "stream": true, "metadata": {}, "provider": "string", "portkey_provider": "string", "portkey_virtual_key": "string" }'

Request

Responses