Troubleshooting¶
This page provides solutions to common issues you might encounter when using VT.ai.
API Key Issues¶
Missing API Keys¶
Problem: VT.ai reports that API keys are missing or invalid.
Solution:
- Check that you've set your API keys using one of these methods:
# Command line
vtai --api-key openai=sk-your-key
# Environment variables
export OPENAI_API_KEY=sk-your-key
- Verify that the API keys are valid by testing them directly with the provider
- Check the
~/.config/vtai/.env
file to ensure keys are saved correctly
Rate Limiting¶
Problem: You're encountering rate limit errors from API providers.
Solution:
- Switch to a different model or provider
- Wait a few minutes before trying again
- Check if your API key has usage restrictions
- Consider upgrading your API tier with the provider
Installation Problems¶
Package Conflicts¶
Problem: Conflicts with existing Python packages during installation.
Solution:
- Use a virtual environment:
python -m venv venv
source venv/bin/activate # Linux/Mac
venv\Scripts\activate # Windows
pip install vtai
- Try installing with uv instead:
uv tool install vtai
- Check for conflicting packages with
pip list
Missing Dependencies¶
Problem: VT.ai reports missing dependencies.
Solution:
- Reinstall with all dependencies:
pip install vtai[all]
- Check if you have enough disk space
- Try updating pip:
pip install --upgrade pip
Performance Issues¶
Slow Responses¶
Problem: VT.ai is taking a long time to respond.
Solution:
- Switch to a faster model (e.g., GPT-o3 Mini, Claude 3 Haiku)
- Check your internet connection
- Reduce query complexity
- Disable dynamic routing if it's causing delays
High Memory Usage¶
Problem: VT.ai is using excessive system memory.
Solution:
- Restart the application
- Use smaller models
- Close other memory-intensive applications
- Increase swap space if possible
Feature-Specific Issues¶
Image Generation Not Working¶
Problem: Image generation commands don't create images.
Solution:
- Ensure you have a valid OpenAI API key with DALL-E access
- Try more explicit prompts like "Generate an image of..."
- Check for error messages in the console output
- Verify that you have sufficient API credits for image generation
TTS Not Working¶
Problem: Text-to-speech functionality isn't working.
Solution:
- Make sure TTS is enabled in settings
- Select a different TTS model
- Check audio output settings on your device
- Verify that you have the necessary API access for TTS features
Vision Analysis Issues¶
Problem: Image analysis doesn't work properly.
Solution:
- Ensure you're using a vision-capable model like GPT-4o
- Check that the image format is supported (JPG, PNG, WebP)
- Try with a different image
- Make sure your query specifically references the image
Chainlit Interface Issues¶
UI Not Loading¶
Problem: The web interface doesn't load properly.
Solution:
- Check if the Chainlit server is running (look for log messages)
- Try a different web browser
- Clear your browser cache
- Check for port conflicts (default is 8000)
Settings Not Saving¶
Problem: Settings changes don't persist between sessions.
Solution:
- Check that you have write permission to
~/.config/vtai/
- Manually edit settings in the configuration files
- Try running with elevated permissions if needed
Getting More Help¶
If you're still experiencing issues:
- Check the GitHub repository for open issues
- Create a new issue with detailed information about your problem
- Join the community discussions for assistance
This page is under construction. More troubleshooting information will be added soon.