> ## Documentation Index
> Fetch the complete documentation index at: https://repr.dev/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# LLM Setup

> Configure Local, Cloud, or BYOK models

Repr relies on Large Language Models (LLMs) to understand your code changes and write stories. You can choose where this processing happens.

## Option 1: Local LLM (Recommended for Privacy)

Run an LLM on your own machine. Requires [Ollama](https://ollama.com).

1. **Install Ollama** and pull a model:
   ```bash theme={null}
   ollama pull llama3.2
   ```

2. **Configure repr**:
   ```bash theme={null}
   repr llm configure
   # Select "Ollama" from the list
   ```

3. **Verify**:
   ```bash theme={null}
   repr llm test
   ```

## Option 2: Bring Your Own Key (BYOK)

Use a commercial model like GPT-4 or Claude without sending data to repr's servers.

1. **Add your key**:
   ```bash theme={null}
   repr llm add openai
   # Enter your sk-... key (stored securely in OS keychain)
   ```

2. **Set as default**:
   ```bash theme={null}
   repr llm use byok:openai
   ```

Supported providers: `openai`, `anthropic`, `groq`, `together`.

## Option 3: Cloud LLM

Let repr manage the models for you. Requires signing in.

1. **Login**:
   ```bash theme={null}
   repr login
   ```

2. **Set as default**:
   ```bash theme={null}
   repr llm use cloud
   ```

<Note>
  Cloud generation requires a subscription plan on repr.dev (free tier available).
</Note>