AI
Chat Completions
OpenAI-compatible chat completion endpoint with streaming support
POST
Request Body
Chat session identifier
Array of message objects
Model identifier
Provider name (e.g., OpenAI, Anthropic, Google)
Sampling temperature (0.0 to 2.0)
Maximum number of tokens to generate
Enable streaming response
Response
Completion ID
Object type (chat.completion)
Unix timestamp of creation
Model used for completion
Array of completion choices
Token usage information

