OpenAI
OpenAI integration via the /v1/chat/completions endpoint.
Configuration
1export OPENAI_API_KEY=sk-...Supported Models
| Model | Input (per 1M) | Output (per 1M) |
|---|---|---|
| gpt-4o | $2.50 | $10.00 |
| gpt-4o-mini | $0.15 | $0.60 |
| gpt-4.1 | $2.00 | $8.00 |
| gpt-4-turbo | $10.00 | $30.00 |
| gpt-3.5-turbo | $0.50 | $1.50 |
Endpoint
Base URL: https://api.openai.com/v1
Requests to OpenAI use the native /v1/chat/completions format — no conversion needed. Authentication uses Authorization: Bearer header.
Model Detection
RelayPlane auto-detects OpenAI models by prefix. Any model starting with these prefixes routes to OpenAI:
gpt-*o1-*,o3-*chatgpt-*text-*,dall-e*,whisper*,tts-*
When routing sends a request from the Anthropic format to OpenAI, RelayPlane automatically converts messages, tools, and tool_choice between formats.