Quickstart
From zero to your first completion in three steps.
- Create an account. Sign up with your email and a password.
- Subscribe. From your dashboard, start the subscription and pay with card or PayPal.
- Create your API key. Click Create API key and store it safely: it is shown only once.
Lost your key? Rotate it from the dashboard. The old key stops working immediately.
Authentication
Every request must include your key as a Bearer token. Keep keys server-side and never ship them in client apps.
Base URL
https://ai.moreunix.com/v1Header
Authorization: Bearer <YOUR_KEY>Format
JSON, OpenAI Chat Completions schema
We recommend exporting it as an environment variable:
Making requests
Use the official OpenAI SDKs or plain HTTP. Only the base URL, key and model ID differ.
Streaming
Set stream: true to receive tokens as they are generated (Server-Sent Events).
Tools & integrations
Any OpenAI-compatible client works. Configure it with these values:
| Tool | Where to configure |
|---|---|
| Open WebUI | Settings → Connections → OpenAI API: base URL + key |
| Continue / Cline | Provider openai with apiBase set to the base URL |
| LangChain | ChatOpenAI(base_url=…, api_key=…, model=…) |
| LlamaIndex | OpenAILike(api_base=…, api_key=…, model=…) |
Models & limits
Use the model ID exactly as shown. Unlimited models have no token cap.
| Model ID | Plans & limits |
|---|
Usage & budgets
Your dashboard shows tokens, requests and spend per model for the last 30 days. Usage appears a few seconds after each request.
- Unlimited models never pause, whatever your volume.
- Limited models include an allowance over a rolling period, counted in tokens (input and output together). When it is used up, that model is paused on your key and resumes automatically as older usage leaves the window.
- Fair use policy: automatic protections against abusive or automated bursts apply to every plan so the service stays fast and reliable for everyone.
Errors
| Status | Meaning | What to do |
|---|---|---|
401 | Missing, invalid or revoked key | Check the header, or rotate your key |
401 | Key blocked | Your subscription is inactive: check billing |
403 | Model not available for your key | Wrong model ID, or the budget of a limited model is used up |
429 | Rate limit exceeded | Retry with exponential backoff |
5xx | Service problem | Retry, and check the status page |