Skip to main content
POST
Create a chat completion

Authorizations

Authorization
string
header
required

App API key (tg_live_…)

Body

application/json
model
string
required

Curated model ID (see Models & pricing).

Example:

"openai/gpt-4o-mini"

messages
object[]
required

OpenAI-format message list.

max_tokens
integer

Output token cap. Defaults to 2048 if omitted (used for cost reservation).

stream
boolean

Stream the response as SSE.

Response

Completion (OpenAI-compatible shape). Streaming responses are SSE.

OpenAI-compatible chat completion object.