v1

latestOpenAPI 3.0.1raw.githubusercontent.com2026-06-022611.2 KB

Chat Completions

Create a chat completion

Creates a model response for the given conversation. Supports streaming via SSE, tool calling, and multi-modal input (text, images, files).

post/api/{workspaceID}/v1/chat/completions

Path parameters

workspaceIDstring required

The workspace ID

Request body

modelstring required

Model ID

streamboolean

Enable SSE streaming

max_tokensinteger

Maximum tokens to generate

max_completion_tokensinteger

Maximum completion tokens (OpenAI-style)

temperaturenumber

Sampling temperature (0-2)

top_pnumber

Nucleus sampling parameter

top_knumber

Top-K sampling parameter

frequency_penaltynumber

Frequency penalty (-2 to 2)

presence_penaltynumber

Presence penalty (-2 to 2)

repetition_penaltynumber

Repetition penalty

seedinteger

Random seed for deterministic output

ninteger

Number of completions to generate

logit_biasobject

Token ID to bias mapping

logprobsboolean

Return log probabilities

top_logprobsinteger

Number of top log probabilities to return

min_pnumber

Minimum probability parameter

top_anumber

Top-A sampling parameter

reasoningobject

Reasoning configuration

reasoning_effortstring

Reasoning effort level

modalitiesstring[]

Output modalities

functionsobject[]

Legacy function definitions

toolsobject[]

Tool definitions for function calling

parallel_tool_callsboolean

Allow parallel tool calls

response_formatobject

Response format specification

predictionobject

Predicted output for latency optimization

{"stackTrail":"components:schemas:ChatCompletionRequest:properties:metadata","oasType":"schema","type":"unknown","description":"Additional metadata"}

storeboolean

Store the completion

service_tierstring

Service tier

userstring

End-user identifier

Example request

{
  "model": "openai/gpt-5"
}

Response

Successful response (non-streaming)

idstring

Log ID

modelstring

Model ID used

object'chat.completion'

createdinteger

All 2 operations