Docs
API reference
Chat Completions

Chat Completions

POSThttps://api.airoutly.com/v1/chat/completions

Generate a model reply to a conversation. Compatible with the OpenAI Chat Completions API, and works with every chat model on Airoutly, including Claude and Gemini.

Authentication

Send your API key as a Bearer token. The OpenAI SDKs do this for you when you pass api_key.

header

Authorization: Bearer $AIROUTLY_API_KEY

Request body

Parameter	Description
`model`stringrequired	Model ID from the models page, such as `gpt-5` or `claude-sonnet-4-5`.
`messages`arrayrequired	The conversation so far. Each message has a `role` (`system`, `user`, `assistant` or `tool`) and `content`. For multimodal models, `content` can be an array of text and `image_url` parts.
`stream`boolean	Return the reply as server-sent events. Recommended for long outputs.
`stream_options`object	Set `{"include_usage": true}` to get token usage in the last chunk of a stream.
`max_tokens`integer	Upper limit on generated tokens. OpenAI reasoning models use `max_completion_tokens` instead.
`temperature`number	Sampling temperature, 0 to 2. Some reasoning models only accept the default.
`tools`array	Functions the model may call. Use with `tool_choice` (`auto`, `none`, `required` or a specific function).
`response_format`object	`{"type": "json_object"}` or a JSON schema, on models that support structured output.

Other OpenAI parameters are passed through to the model. Whether they take effect depends on the model.

Example request

curl https://api.airoutly.com/v1/chat/completions \
  -H "Authorization: Bearer $AIROUTLY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ]
  }'

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.airoutly.com/v1",
    api_key=os.environ["AIROUTLY_API_KEY"],
)
reply = client.chat.completions.create(
    model="gpt-5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"},
    ],
)
print(reply.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.airoutly.com/v1",
  apiKey: process.env.AIROUTLY_API_KEY,
});
const reply = await client.chat.completions.create({
  model: "gpt-5",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "Hello!" },
  ],
});
console.log(reply.choices[0].message.content);

Response

200 OK

{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1762300800,
  "model": "gpt-5",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "Hello! How can I help?" },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 19, "completion_tokens": 8, "total_tokens": 27 }
}

You are billed for the usage shown here, at the model’s price on the models page.

Streaming

With "stream": true, the reply arrives as a series of data: lines and ends with data: [DONE]. Add stream_options to get usage in the last chunk.

text/event-stream

data: {"object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hel"}}]}

data: {"object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"lo!"}}]}

data: {"object":"chat.completion.chunk","choices":[],"usage":{"prompt_tokens":19,"completion_tokens":8,"total_tokens":27}}

data: [DONE]

Errors

Errors use the OpenAI error shape. Rely on the HTTP status and code; the message ends with a request ID to quote when you contact support. See error codes for what each one means.

503

{
  "error": {
    "message": "No available channel for model gpt-6 (request id: 20261105...)",
    "type": "new_api_error",
    "code": "model_not_found"
  }
}