Create a chat completion using the specified model. Supports multi-turn conversation, streaming, and a variety of generation parameters.
All endpoints require Authorization: Bearer YOUR_API_KEY.
Model name (e.g. gpt-4o, gpt-5, claude-opus-4-1-20250805).
List of conversation messages.
Controls output randomness (0-2). Default: 1.0
Maximum number of tokens to generate.
Whether to use streaming output (SSE). Default: false
Nucleus sampling parameter (0-1). Default: 1.0
Frequency penalty (-2.0 to 2.0). Default: 0
Presence penalty (-2.0 to 2.0). Default: 0
Stop sequences (up to 4).
Number of completions to generate. Default: 1