document local model setup

README.md

@@ -29,6 +29,8 @@ The one-line installer fetches the latest tagged release. To pin a version, pass
 
 If you install via `pnpm` or `bun` instead of the standalone bundle, Feynman requires Node.js `20.19.0` or newer.
 
+Local models are supported through the custom-provider flow. For Ollama, run `feynman setup`, choose `Custom provider (baseUrl + API key)`, use `openai-completions`, and point it at `http://localhost:11434/v1`.
+
 ### Skills Only
 
 If you want just the research skills without the full terminal app:

website/src/content/docs/getting-started/setup.md

@@ -42,6 +42,33 @@ For API key providers, you are prompted to paste your key directly:
 
 Keys are encrypted at rest and never sent anywhere except the provider's API endpoint.
 
+### Local models: Ollama, LM Studio, vLLM
+
+If you want to use a model running locally, choose the API-key flow and then select:
+
+```text
+Custom provider (baseUrl + API key)
+```
+
+For Ollama, the typical settings are:
+
+```text
+API mode: openai-completions
+Base URL: http://localhost:11434/v1
+Authorization header: No
+Model ids: llama3.1:8b
+API key: local
+```
+
+That same custom-provider flow also works for other OpenAI-compatible local servers such as LM Studio or vLLM. After saving the provider, run:
+
+```bash
+feynman model list
+feynman model set <provider>/<model-id>
+```
+
+to confirm the local model is available and make it the default.
+
 ## Stage 3: Optional packages
 
 Feynman's core ships with the essentials, but some features require additional packages. The wizard asks if you want to install optional presets: