Frequently Asked Questions

What is Privacy LLama?

Privacy LLama is a privacy-first web application that lets you chat with any AI model using your own API keys. It's perfect for testing different models, avoiding vendor lock-in, and maintaining complete privacy.

Is Privacy LLama free?

Yes! Privacy LLama is 100% free to use. However, you'll need your own API key from a provider like OpenAI or OpenRouter, which may have their own costs based on your usage.

Do I need to create an account?

No! Privacy LLama requires no account, no signup, and no personal information. Just bring your API key and start chatting.

Why should I use this instead of ChatGPT or Claude?

Great question! Here's why Privacy LLama might be better for you:

• Test any model: Not limited to one provider's models
• Privacy: No accounts, no tracking, no data collection
• Flexibility: Use OpenAI, Claude, Gemini, Llama, and more from one interface
• Cost control: Pay only for what you use via API (often cheaper than subscriptions)
• No rate limits: Subject only to your API provider's limits

Where do I get an API key?

You can get API keys from several providers:

• OpenAI: platform.openai.com/api-keys
• OpenRouter: openrouter.ai (access to 100+ models)
• Azure OpenAI: Via Azure Portal (enterprise)
• Local LLMs: LM Studio, Ollama with OpenAI compatibility

Is my API key safe?

Yes! Your API key is stored only in your browser's localStorage. It never leaves your device, and we have no servers to collect it. Your browser communicates directly with your chosen API provider.

Can I use multiple API keys?

Absolutely! You can save multiple API keys and switch between them instantly. Great for comparing different providers or keeping personal and work keys separate.

What's the Base URL?

The Base URL is the API endpoint. Common ones:

• OpenAI: https://api.openai.com/v1
• OpenRouter: https://openrouter.ai/api/v1
• Local LLMs: Usually http://localhost:1234/v1 or similar

What models can I use?

Any model supported by your API provider! Popular options include:

• OpenAI: GPT-5, GPT-4o, GPT-5 Pro
• Anthropic (via OpenRouter): Claude 4.5 Opus, Sonnet, Haiku
• Google (via OpenRouter): Gemini Pro
• Meta (via OpenRouter): Llama 2, Llama 3
• Mistral AI: Mistral, Mixtral models
• And many more: 100+ models via OpenRouter

How do I use a model that's not in the list?

Click "Custom Model" in the model selector and enter the exact model name. Make sure it's supported by your API provider.

Can I stop a response mid-generation?

Yes! While the AI is generating a response, the Send button changes to a red "Stop" button. Click it to stop generation immediately.

Does it support streaming?

Yes! Responses stream in real-time, just like ChatGPT. You'll see the AI typing with animated dots, then the response appears word by word.

Is this really private?

Yes! Privacy LLama runs entirely in your browser. We don't have any servers, databases, or analytics. Your API keys and conversations never leave your device. See our Privacy Policy for details.

Is it GDPR compliant?

100% compliant! We don't collect any data, so there's nothing to be non-compliant about. You also have full control with export/import and the "Clear All Data" button.

Where is my data stored?

All data (API keys, conversations, settings) is stored in your browser's localStorage. It never leaves your device and is only accessible to you on that browser.

What happens if I clear my browser data?

Your API keys and conversations will be deleted (they're stored in localStorage). Use the Export feature to backup important conversations before clearing browser data!

Can I backup my conversations?

Yes! Use the Export button to download any conversation as a JSON file. You can import it back later using the Import button.

How do I delete all my data?

Click the "🗑️ Clear All Data (GDPR)" button at the bottom of the API Keys section. This will permanently delete all API keys and conversations from your browser.

Does it support markdown and code?

Yes! All AI responses support full markdown formatting with syntax-highlighted code blocks. You can also copy any message with the copy button that appears on hover.

Can I save multiple conversations?

Yes! Create as many conversations as you want. They're automatically saved and you can switch between them using the "Show Conversations" button.

Can I share conversations?

Export the conversation as a JSON file and share the file. The recipient can import it into their Privacy LLama instance (they'll need their own API key to continue the conversation).

What browsers are supported?

Privacy LLama works in all modern browsers (Chrome, Firefox, Safari, Edge). You need JavaScript enabled and localStorage support.

Does it work offline?

The interface loads offline after first visit, but you need an internet connection to communicate with AI providers' APIs.

Can I use this with local LLMs?

Yes! If you're running a local LLM server with OpenAI-compatible API (like LM Studio or Ollama), just enter your local server's URL as the Base URL (e.g., http://localhost:1234/v1).

Is the code open source?

The implementation follows standard web practices. You can inspect the code in your browser's developer tools.

Models aren't loading

Check that your API key is valid and has the correct permissions. If models still don't load, use "Custom Model" to enter the model name manually.

I'm getting API errors

Common causes:

• Invalid or expired API key
• Insufficient credits/quota on your API account
• Wrong base URL for your provider
• Model name not supported by your provider

Check your browser's console (F12) for detailed error messages.

Streaming is slow or not working

This is usually due to your API provider's response time or rate limits, not Privacy LLama. Try a different model or check your API provider's status.