Frequently asked questions
Common questions about the ReadyServer LLM API.
Is the LLM API really free?
Yes. It's included with your ReadyServer VPS at no extra cost. There's no per-token billing; usage is governed by our fair use policy.
Who can use it?
ReadyServer VPS customers. The API is only reachable from ReadyServer VPS over our private network.
Every ReadyServer VPS plan includes access, with nothing extra to set up.
Why is there no API key?
Access is controlled by the network: only ReadyServer VPS can reach the endpoint, so there's no key to create, rotate or leak. Some tools insist on a key anyway. Enter any placeholder, such as not-needed.
Can I use it from my own computer or another cloud?
Not directly. The endpoint isn't on the public internet, so run your application on your ReadyServer VPS.
You can reach the API from your own computer through an SSH tunnel to your VPS for personal development, but don't share access with others.
Which models are available?
We currently serve Qwen3.6-27B and Qwen3.8-27B, with more models planned. See Models for details.
Which SDKs and tools work with it?
Anything that supports an OpenAI-compatible API with a custom base URL, including the official OpenAI SDKs for Python and Node.js, and many agent frameworks, chat interfaces and coding agents. Tools built for the Anthropic Messages API work too. See the quickstart and API reference.
Are my prompts and responses logged or used for training?
No. We don't log prompts or responses, so they can't be used for training.
What happens when the service is busy?
Requests are queued, and customers who have used fewer tokens are served first, so responses can take longer to start at peak times. The fair use policy explains the details.
Is there an SLA for the LLM API?
The LLM API is provided on a best-effort basis, without an SLA.
Will more models be added? Can I request one?
Yes, more models are planned. Once the community forum opens you'll be able to request models there. Until then, contact support.
Where can I get help?
Start with Troubleshooting. If you're still stuck, contact ReadyServer support.