Direct Model Access

Chat Completions

Direct LLM access with automatic provider fallback.

Overview

POST /v1/chat/completions sends a prompt directly to an LLM using your connected provider API keys. This is the simplest way to use Redeo as a drop-in replacement for OpenAI — same request shape, same response shape.

Redeo routes your request through the provider connections you've configured (OpenAI, Anthropic, DeepInfra, etc.) with automatic fallback if a provider fails.

Request Body

Field	Type	Required	Description
`model`	`string`	Yes	Model to use (e.g. `gpt-4o`, `claude-sonnet-4-20250514`, `deepseek-r1`). Must be supported by one of your configured providers.
`messages`	`array`	Yes	Array of message objects with `role` (`system`, `user`, `assistant`) and `content` (string or content parts). The last `user` message is used as the prompt.
`temperature`	`number`	No	Sampling temperature. Passed through to the provider.
`max_tokens`	`integer`	No	Maximum tokens to generate. Passed through to the provider.
`stream`	`boolean`	No	Not yet supported. Must be omitted or `false`.

Response Shape

The response is a standard OpenAI chat completion object:

json

{
  "id": "redeo-chat-<uuid>",
  "object": "chat.completion",
  "created": 1716234567,
  "model": "gpt-4o",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The generated response..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 42,
    "completion_tokens": 128,
    "total_tokens": 170
  }
}

Example

bash

curl https://api.redeo.ai/v1/chat/completions \
  -H "Authorization: Bearer $REDEO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

This uses your highest-priority provider connection that supports gpt-4o. If that provider fails, Redeo falls back to the next available provider automatically.