Connect a model for BG3 modding

Mythic Loop runs on your Windows desktop and needs a model to process text. You can pick a cloud provider, use a local PC model, or rely on the built-in option. Each choice changes privacy and speed.

Choose your model source

Cloud provider

Send your prompts to a cloud AI provider through a standard API endpoint. Stronger cloud models are available, but prompt text and file text go to that provider.

Local PC model

Run a model directly on your Windows PC using available hardware. This keeps all data inside your computer during processing. No network traffic occurs for the core work. You can edit the model settings if you have advanced knowledge. This is the best choice for strict privacy needs.

Built-in model

Use the built-in model that comes with the application. It is ready to go without extra configuration steps. This option simplifies the setup process for beginners. You get immediate access to the agent features.

Test your connection

The Test button checks an endpoint and shows response time and the context window.

Select the right model size

Use a bigger model for refactors or complex logic. Smaller models work well for small edits. You can send a hard step to a bigger model using run_on_model. This gives you flexibility for different tasks. You can switch models as needed during your session.

Quality varies between different model providers. Always review the diff before accepting changes. Use Ask first to pause before actions that change something. This gives you control over the process. You can verify the output before it is applied.