One consistent API.

Reference for authentication, chat completions, streaming events, model discovery, rate limits, and errors.

API VERSION 2026-08BASE URL /V1

Authentication

Authenticate every request with a project API key in the Authorization header.

HEADERAuthorization: Bearer $STANDARD_API_KEYREQUIRED

KEY SCOPE

Keys inherit project limits and can be individually revoked. Use a separate key for each production service.

POST/v1/chat/completionsJSON

Generate a response from a text or multimodal model. Set stream to true for incremental server-sent events.

modelstring · required

The model ID to use for generation.

messagesarray · required

Conversation messages in system, user, and assistant roles.

streamboolean

Returns incremental SSE events when true.

max_tokensinteger

Maximum number of output tokens to generate.

temperaturenumber

Sampling temperature between 0 and 2.

toolsarray

Functions the model may call during generation.

Streaming responses use server-sent events. Each event contains a partial delta, followed by a final [DONE] marker.

data: {"choices":[{"delta":{"content":"Standard Thinking"}}]}

data: {"choices":[{"delta":{"content":" is ready."}}]}

data: [DONE]

GET/v1/modelsJSON

List models available to the current project, including model ID, family, status, and context length.

Limits are applied per project and model. Response headers report remaining requests and tokens for the current window.

x-ratelimit-limit-requestsheader

Maximum requests in the current window.

x-ratelimit-remaining-requestsheader

Requests available before reset.

x-ratelimit-reset-requestsheader

Time until the request window resets.

400bad_request

The request body or parameter value is invalid.

401unauthorized

The API key is missing, invalid, or revoked.

429rate_limit

The project exceeded its request or token limit.

500server_error

An unexpected error occurred. Retry with the Request ID.