OpenAI-compatible access to chat models, images, video and audio. Tasks for long generations, error handling and spending reports are built in: in your program you only change the address and the key.
curl https://api.6io.io/v1/chat/completions \
-H "Authorization: Bearer sixio_sk_..." \
-H "Content-Type: application/json" \
-d '{
"model": "op/openai/gpt-4o-mini",
"messages": [{"role": "user", "content": "Hello!"}],
"max_tokens": 100
}'from openai import OpenAI
client = OpenAI(base_url="https://api.6io.io/v1", api_key="sixio_sk_...")
reply = client.chat.completions.create(
model="op/openai/gpt-4o-mini",
messages=[{"role": "user", "content": "Hello!"}],
max_tokens=100,
)
print(reply.choices[0].message.content)Official SDKs work unchanged: just change the address and the key. Streaming responses (SSE) are supported.
Chat models of the op and vn groups; images, video and audio from the fal catalog and the 6io (sixio) models.
A separate balance in credits, like on 6io. An estimate is reserved before the request, the actual cost is charged and the rest of the reserve is returned.
POST /v1/jobs for long requests, a notification to your address instead of polling. Idempotency-Key: a retry never creates a second generation or charges twice.
Every request shows its state and the charge: in the console and through GET /v1/requests/{id}.
Several keys, each with its own expiry. Limits apply to the whole account, not to a key.
A model name starts with its group prefix. Chat answers as a stream: the answer arrives in the same connection. Images, video and audio run as a queue: the request returns an ID at once, and you collect the result by it or get a notification at your address. Models and prices are in the console, on the Models tab.
op/ | Chat models through OpenRouter · stream |
vn/ | Venice chat models · stream |
fal/ | Images, video and audio from the fal catalog · queue |
sixio/ | Images and video connected in 6io · queue |
The price of each model is given in credits: in the console on the Models tab and in the API catalogs (described in the documentation). The actual cost of a request is charged; you choose the top-up amount.
With 6io site credits: in the console transfer the number you need, one site credit gives one API credit. The transfer is available after the first payment on the site, and the amount cannot be moved back to site credits.
We do not store finished results: file links are valid for a limited time, so download them right away. Details are in the documentation.
You are. The API terms of use describe what 6io is responsible for and what is left to the user and the model providers.
The console shows the balance, keys, request history and spending, and the playground sends the first request right from the browser.
Open the console