Connect your AI model
SourceLens runs tasks using an AI model you choose. You select a cloud provider, a local model on your PC, or the built-in option. Each choice changes how data moves and what privacy you keep.
Choose your model
Cloud AI provider
Your prompts and file text the agent reads go to that provider. The app shows response time and context window details.
Local PC model
Run a model locally on your PC. Your prompts and file contents stay on the PC; web tools and update checks still use the network. This is the best choice for strict privacy. You control the model file. The agent reads your inventory and writes diffs without sending data anywhere.
Built-in model
Use the built-in model for simple tasks. It is ready to use without your own endpoint or API key. The agent runs commands and reads files.
Test your connection
Press the Test button to check an endpoint. The app shows response time and the context window.
Pick the right model
Use a bigger model for refactors or complex logic. Send a hard step to a bigger model with run_on_model. Use a smaller model for small edits. This saves tokens. The agent splits tasks. You review the plan. The tool runs commands. You see the result.
Quality varies between models. Review the diff in Ask first. The agent shows changes before you accept. You can edit the plan. Shell runs terminal commands. Success is judged by the real exit code. The agent reads files. You control the workflow. You decide what to keep.