# Frequently Asked Questions

## Which provider should I choose?

**For free usage:**

- **OpenRouter** \- Best free option, access to Llama 3.3, Qwen Coder
- **Ollama** \- If you have a good GPU and want privacy

**For best quality:**

- **Claude Sonnet 4** \- Excellent for coding
- **GPT-4o** \- Most capable overall

## Is my API key secure?

Yes! Your API key is:

- Stored locally in `~/.coderrr/config.json`
- Sent directly to your chosen provider
- Never logged or stored on our servers

## Can I use Coderrr without an API key?

Yes! Use either:

- **Ollama** \- Run models locally (no key needed)
- **OpenRouter free tier** \- Sign up for a free API key

## What models are free?

**OpenRouter Free Models:**

- meta-llama/llama-3.3-70b-instruct:free
- qwen/qwen-2.5-coder-32b-instruct:free
- deepseek/deepseek-chat-v3-0324:free

**Ollama (all local models are free):**

- qwen2.5-coder:7b
- llama3.1:8b
- codellama:7b

## How do I switch providers?

```
coderrr config
```

Select a new provider and it will overwrite your previous config.

## Does Coderrr work offline?

Yes! With Ollama, you can run models completely offline after downloading them:

```
ollama pull qwen2.5-coder:7b
coderrr config  # Select Ollama
coderrr  # Works offline!
```