Choose your AI model
SourceHarbor connects to an external cloud provider via a standard API endpoint, a model running locally on your PC, or the built-in model. Each option changes privacy, speed, and cost. You pick the one that fits your workflow.
Three ways to run intelligence
Cloud AI provider
Connect to a cloud AI provider through a standard API endpoint. The prompt and file text travel to that provider, so your code leaves the machine. Privacy depends on the provider’s data retention policy. You may pay for usage.
Local PC model
With a local model nothing leaves the PC.
Built-in model
Use the built-in model for simple edits and quick checks. This option requires no external configuration or API keys. It is ready to use without your own endpoint. The model handles routine tasks.
Verify your connection
Press the Test button to check your endpoint. The tool shows the response time and context window size. This confirms the model is reachable and ready. You can adjust settings before starting work.
Select the right model size
Use a bigger model for complex refactors and hard logic steps. Send these steps to a larger model using run_on_model.
Quality varies between models and runs. Always review the diff in Ask first mode. This mode pauses before actions that change something. You see the exact file changes before accepting. This prevents unwanted modifications to your code. Trust the diff to confirm the agent’s intent.