Choose the AI model that fits your privacy and speed needs

Java Trail connects to a cloud provider, a local PC model, or uses the built-in option. You pick the endpoint, test it, and keep your data on your own Windows machine when using a local model; with a cloud provider, prompts go to the provider.

Three ways to run the AI brain

Use a cloud AI provider

Send prompts to a standard API endpoint on the internet. The model processes your code and returns suggestions. Your file text leaves the PC, but you control which provider receives it. This works well when you need answers for complex logic.

Run a model locally on your PC

Keep all data inside your Windows machine. With a local model, prompts stay on the PC; optional network features (web search, update checks, phone control) still exist. You can run this alongside your Java compiler.

Use the built-in model

Ready to use without your own endpoint.

Check your connection with the Test button

Press the Test button to verify your chosen endpoint. The tool shows the response time and the context window size. This confirms that the model is reachable.

Pick the right model for the task

Use a bigger model for complex refactors or deep logic. Send a hard step to a bigger model using the run_on_model tool. Smaller models work well for small edits. You can switch models depending on the difficulty of the Java code you are writing.

Quality varies between models and tasks. Always review the diff before accepting. Use Ask first mode to pause before changes. This lets you verify the code logic. You stay in control of every file edit.