Connect an AI model for your CS2 code

Nocode Atelier works with a cloud AI provider, a model running on your PC, or the built-in model. You choose one, press Test and see the response time. The choice decides where your prompts go and how much text the agent can read at once.

Three ways to connect a model

A cloud AI provider

Use a standard API endpoint from a cloud AI provider. It suits large plugins or long config sets. Your prompts go to that endpoint, and the provider may charge you separately. Secrets in tool output are redacted before they reach the model.

A model on your PC

A model running locally on your PC suits private projects. With a local model nothing leaves the machine. Smaller models may struggle with big refactors, so keep each task small and read every diff before you accept it.

The built-in model

The built-in model is part of the app, so you can try a task before you set up an endpoint. Use the Test button to see its context window, keep tasks small and review every diff before you test the result.

Check the connection with Test

The Test button checks your endpoint before you start. It shows the response time and the context window, which is how much text the model can hold at once. Read both numbers before you ask the agent to open a long server log. If the test fails, fix the endpoint settings first.

Pick a model for the job

Use a bigger model for a refactor or tricky plugin logic, and a smaller one for small edits to a .cfg file. When one step is too hard, run_on_model sends just that step to a bigger model while the rest of the chat stays where it is.

Quality varies between models. Read every diff, keep Ask first switched on for new projects, and test the result on a local server before real players join.