​
Environment Setup

Auto-Browse supports multiple Language Model (LLM) providers to power its AI capabilities. This guide covers the setup for each supported provider.

​
Supported Providers

​
OpenAI (Default Provider)

  1. Get your OpenAI API key:
    • Go to OpenAI’s platform
    • Create an account or sign in
    • Navigate to API keys
    • Create a new API key
  2. Configure your environment: 
OPENAI_API_KEY=your_openai_api_key_here
LLM_PROVIDER=openai  # Optional, defaults to openai
AUTOBROWSE_LLM_MODEL=gpt-4o-mini  # Optional, defaults to gpt-4o-mini

​
Model Options

OpenAI models supported:
  • gpt-4o-mini (default)
  • Compatible GPT-4 models

​
Google AI (Gemini)

  1. Get your Google AI API key:
  2. Configure your environment: 
GOOGLE_API_KEY=your_google_key_here
LLM_PROVIDER=google
AUTOBROWSE_LLM_MODEL=gemini-2.0-flash-lite

​
Model Options

  • gemini-2.0-flash-lite
  • Other Gemini models

​
Azure OpenAI

  1. Set up Azure OpenAI:
    • Create an Azure account
    • Set up Azure OpenAI service
    • Deploy a model
    • Get your API details
  2. Configure your environment: 
AZURE_OPENAI_API_KEY=your_azure_key_here
AZURE_OPENAI_ENDPOINT=https://your-endpoint.openai.azure.com/
AZURE_OPENAI_API_VERSION=2024-12-01-preview
AZURE_OPENAI_API_DEPLOYMENT_NAME=your-deployment-name
LLM_PROVIDER=azure

​
Anthropic Claude

  1. Get your Anthropic API key:
  2. Configure your environment: 
ANTHROPIC_API_KEY=your_anthropic_key_here
LLM_PROVIDER=anthropic
AUTOBROWSE_LLM_MODEL=claude-3

​
Google Vertex AI

  1. Set up Google Cloud:
    • Create a Google Cloud project
    • Enable Vertex AI API
    • Create service account credentials
  2. Configure your environment: 
GOOGLE_APPLICATION_CREDENTIALS=path/to/credentials.json
LLM_PROVIDER=vertex

​
Ollama (Local Deployment)

  1. Install Ollama:
    • Follow installation instructions at Ollama.ai
    • Start the Ollama service
  2. Configure your environment: 
LLM_PROVIDER=ollama
AUTOBROWSE_LLM_MODEL=llama2
BASE_URL=http://localhost:11434  # Optional, this is the default

​
Best Practices

  1. API Key Security
    • Never commit API keys to version control
    • Use environment variables or secret management
    • Rotate keys periodically
  2. Model Selection
    • Start with default model for basic operations
    • Test different models for specific use cases
    • Consider cost vs. performance tradeoffs
  3. Error Handling
    • Implement proper error handling
    • Monitor API rate limits
    • Set appropriate timeouts

​
Troubleshooting

Common issues and solutions:
  1. API Key Issues
Error: Authentication failed
Solution: Verify API key is correct and has proper permissions
  1. Model Availability
Error: Model not available
Solution: Confirm model name and availability in your region
  1. Rate Limits
Error: Rate limit exceeded
Solution: Implement backoff strategy or upgrade API tier

​
Next Steps