Create a chat completion
Generates a model response for a conversation. Supports streaming (stream: true, server-sent events) and non-streaming responses. When streaming, each frame is a data: line holding one JSON completion chunk, the chunk carrying usage has an empty choices array, and the stream ends with the literal data: [DONE]. If the request fails after streaming has started, the stream ends without [DONE], usually on a final data: line holding an error object. OpenAI-compatible: the supported fields are documented here, and the wider OpenAI SDK parameter set is accepted so that existing OpenAI code runs unmodified. Unrecognized fields are rejected with a 400 (unknown_parameter).
Authentication
Your ASI:One API key, sent as Authorization: Bearer <api_key>.
Headers
Groups related requests into one session. Required when using planner_mode, which uses it to resume a paused plan.
Request
The ASI:One model to use.
If true, partial deltas are streamed as server-sent events.
Controls which (if any) tool is called.
Constrains the reply’s format. A json_schema that is invalid returns a 400 (invalid_json_schema) whose param points at the offending part of the schema. See the Structured Data guide.
ASI:One extension. Turns reasoning on or off on asi1-mini. The other models always reason; there an explicit false selects the shallowest depth. See the Reasoning guide.
ASI:One extension. The older way to size reasoning. On asi1 and asi1-ultra the value selects a depth rather than capping tokens, and reasoning_effort wins when both are sent. See the Reasoning guide.
How deeply the model reasons. minimal and xhigh are accepted as well, matching OpenAI’s wider scale: minimal behaves as low, xhigh as high. No effect on asi1-mini. See the Reasoning guide.
ASI:One extension. Routes the request to the planner, which breaks it into steps and runs them against tools and Agentverse agents. Available on asi1, asi1-ultra and asi1-mini. See the Planner Mode guide.
ASI:One extension. Agentverse agent addresses the planner may use without discovering them first. Used with planner_mode.
Number of choices to generate. Only 1 is supported.
Lets the model search the web for this request. Sending the object turns search on. Not available with planner_mode. See the Web Search guide.
Response headers
Correlation id for this request, present on every response: the caller-supplied X-Request-Id when one was sent, otherwise minted by the API. Backends log it and traces key on it, so cite it when reporting an issue.
Response
A chat completion object, or an SSE stream of completion chunks when stream is true.
Unix timestamp (seconds) of creation.