๐ง Custom API (Optional)
๐ก Recommended for most users: Use HuBrowser membership for the best experience - faster, more reliable, and better value!
Advanced option: For users who need maximum control, compliance requirements, or want to experiment, you can configure HuBrowser Extension to use your own custom AI API.
Click the "API Config" button at the top of the extension to configure your own API keys.
๐ฏ Why Use Your Own API?
โ Full Control
Use any AI model provider - ChatGPT, Claude, Gemini, DeepSeek, or any OpenAI-compatible API.
๐Maximum Privacy
Run entirely locally on-device with edge AI models - no internet connection required. You can inspect the network traffic to confirm requests go directly to your chosen provider with no requests to our servers.
๐ฐCost Transparency
- Higher cost than HuBrowser membership due to our custom optimizations and bulk pricing agreements
- Still cheaper than using the API with other agentic AI solutions thanks to our efficiency optimizations
- Full visibility into your API usage and costs
๐งชExperimentation & Comparison
Objectively compare different AI models, providers, and agent solutions for your specific use cases.
๐ Supported Configurations
You can use any of these AI providers:
- OpenAI API - GPT-4, GPT-4o, GPT-3.5, GPT-4o-mini
- Anthropic Claude API - Claude 3.5 Sonnet, Claude 3 Opus, Claude 3 Haiku
- Google Gemini API - Gemini Pro, Gemini Ultra
- DeepSeek API - DeepSeek Chat, DeepSeek Coder
- Any OpenAI-compatible API endpoint - Groq, Together AI, Perplexity, etc.
- Local AI models - Ollama, LM Studio, Jan, etc.
๐จ How to Configure
Step 1: Get Your API Key
Obtain an API key from your chosen provider:
- OpenAI: platform.openai.com
- Anthropic: console.anthropic.com
- Google: aistudio.google.com
- DeepSeek: platform.deepseek.com
- Any other provider's API key page
Step 2: Open API Config
- Click the HuBrowser Extension icon in your browser toolbar
- Click the "API Config" button at the top
- Enter your API configuration
Step 3: Enter Your Settings
Fill in the required fields:
- API Endpoint - The base URL for the API (e.g.,
https://api.openai.com/v1) - API Key - Your personal API key
- Model Name - The specific model to use
Step 4: Save and Test
Click Save and try a command to verify everything works!
๐ก Best Practices
Choose the Newer Models
- Complex tasks (automation, coding) โ Use larger models like GPT-5 or Claude Sonnet
- Privacy-critical โ Use local models with Ollama
Monitor Your Usage
Keep track of API costs in your provider's dashboard. Our optimizations help reduce token usage, but costs can add up with frequent use.
Rate Limits
Be aware of your API provider's rate limits. If you hit limits, consider:
- Switching to HuBrowser membership (we have higher limits)
- Upgrading your API tier
- Spacing out commands
๐ Privacy & Security
Your Data Stays Private
When using your own API:
- Requests go directly to your configured endpoint
- No data is sent to HuBrowser servers
- You can verify this by monitoring network traffic
Verifiable Security
Our extension is open source - you can:
- Review the code on GitHub
- Inspect network requests in browser DevTools
- Confirm no unexpected data transmission
Data Residency Compliance
Perfect for users who need to comply with:
- GDPR requirements
- Corporate data policies
- Government regulations
- Industry-specific compliance
๐ Comparison: Custom API vs HuBrowser Membership
| Feature | Custom API | HuBrowser Membership |
|---|---|---|
| Privacy | ๐ข Maximum (local/direct) | ๐ก Standard (encrypted) |
| Speed | ๐ก Depends on provider | ๐ข Optimized infrastructure |
| Cost | ๐ก Pay per token | ๐ข Fixed monthly price |
| Setup | ๐ก Manual configuration | ๐ข Instant, no setup |
| Model Choice | ๐ข Any model | ๐ก Curated selection |
| Support | ๐ก Self-service | ๐ข Priority support |
โ Troubleshooting
"Invalid API Key" Error
- Double-check your API key is copied correctly
- Verify the key is active in your provider's dashboard
- Make sure you have billing enabled (for paid APIs)
"Model Not Found" Error
- Confirm the model name matches exactly (case-sensitive)
- Check your API tier has access to that model
- For local models, ensure the model is pulled (
ollama list)
Slow Response Times
- Check your internet connection (for cloud APIs)
- Try a smaller/faster model
- For local models, ensure sufficient RAM available
Connection Timeout
- Verify the API endpoint URL is correct
- Check firewall settings aren't blocking requests
- For Ollama, ensure it's running (
ollama serve)
๐ค Need Help?
- Documentation: Check provider-specific docs for API details
- Support: Contact us for enterprise custom AI deployment
Ready to take control of your AI? Configure your own API and experience HuBrowser Extension on your terms! ๐