Create a chat completion | AssemblyAI

To use our EU server for LLM Gateway, replace llm-gateway.assemblyai.com with llm-gateway.eu.assemblyai.com.

Generates a response from a model given a prompt or a series of messages.

<llms-only> > For the complete documentation index, see [llms.txt](https://www.assemblyai.com/docs/llms.txt) </llms-only> <Note>To use our EU server for LLM Gateway, replace `llm-gateway.assemblyai.com` with `llm-gateway.eu.assemblyai.com`.</Note> Generates a response from a model given a prompt or a series of messages.

Authentication

Authorizationstring

API Key authentication via header

Request

Request body for creating a chat completion.

modelstringRequired

The ID of the model to use for this request. See LLM Gateway Overview for available models.

messageslist of objectsOptional

A list of messages comprising the conversation so far.

promptstringOptional

A simple string prompt. The API will automatically convert this into a user message.

max_tokensintegerOptional>=1Defaults to 1000

The maximum number of tokens to generate in the completion. Default is 1000.

temperaturedoubleOptional

Controls randomness. Lower values produce more deterministic results.

streambooleanOptionalDefaults to false

When true, responses are streamed as server-sent events (SSE). Supported on OpenAI models only.

toolslist of objectsOptional

A list of tools the model may call.

tool_choiceenum or objectOptional

Controls which (if any) function is called by the model.

response_formatobjectOptional

Specifies the format of the model’s response. Use this to constrain the model to output valid JSON matching a schema. Supported by OpenAI (GPT-4.1, GPT-5.x), Gemini, and Claude models. Not supported by gpt-oss models.

fallbackslist of objectsOptional

An array of fallback objects. Each object must include a model and can optionally override any field from the original request. If the primary model fails, the LLM Gateway tries each fallback in order until one succeeds. See Specify fallback models for more details.

fallback_configobjectOptional

Configuration for fallback behavior, including retry and depth settings. See Specify fallback models for more details.

Response

Successful response containing the model's choices.

request_idstringformat: "uuid"

choiceslist of objects

requestobject

A copy of the original request, excluding prompt and messages.

usageobject

http_status_codeinteger

The HTTP status code of the response

response_timeinteger

The response time in nanoseconds

llm_status_codeinteger

The status code from the LLM provider

1	curl -X POST https://llm-gateway.assemblyai.com/v1/chat/completions \
2	-H "Authorization: <apiKey>" \
3	-H "Content-Type: application/json" \
4	-d '{
5	"model": "claude-sonnet-4-6",
6	"prompt": "Write a haiku about coding",
7	"max_tokens": 50,
8	"temperature": 0.5
9	}'

1	{
2	"request_id": "d4f1e7a2-9c3b-4f5a-8e2d-1a2b3c4d5e6f",
3	"choices": [
4	{
5	"message": {
6	"role": "assistant",
7	"content": "Silent lines of code,\nLogic flows like gentle streams,\nDreams in syntax bloom."
8	},
9	"finish_reason": "stop",
10	"tool_calls": [
11	{
12	"id": "fc1234567890",
13	"type": "function",
14	"function": {
15	"name": "generate_haiku",
16	"arguments": "{\"topic\":\"coding\"}"
17	}
18	}
19	]
20	}
21	],
22	"request": {
23	"model": "claude-sonnet-4-6",
24	"max_tokens": 50,
25	"temperature": 0.5,
26	"tools": [
27	{
28	"type": "function",
29	"function": {
30	"name": "generate_haiku",
31	"parameters": {},
32	"description": "Generates a haiku poem based on a given topic."
33	}
34	}
35	],
36	"tool_choice": "none"
37	},
38	"usage": {
39	"input_tokens": 12,
40	"output_tokens": 45,
41	"total_tokens": 57
42	},
43	"http_status_code": 200,
44	"response_time": 275510459,
45	"llm_status_code": 200
46	}