Automatically tag your photos with AI-powered captions, descriptions, and keywords using Google Gemini AI or local Ollama models.
- Smart Captions: Generate concise, descriptive captions (1-2 sentences)
- Detailed Descriptions: Create comprehensive descriptions (2-3 sentences)
- Intelligent Keywords: Extract relevant keywords automatically
- Usage Instructions: Get AI suggestions for photo editing and usage
- Copyright Detection: Identify visible copyright or attribution information
- Location Recognition: Detect identifiable landmarks and locations
- Complete IPTC Support: Save all metadata to industry-standard IPTC fields
- Smart Keyword Storage: Keywords saved to Lightroom Keywords (automatically included in IPTC on export)
- Configurable Options: Choose which metadata fields to save
- Batch Processing: Analyze multiple photos efficiently with rate limiting
- Export Functionality: Export analysis results to CSV for external processing
- Custom Prompts: Create your own AI analysis prompts
- Batch Configuration: Adjust batch size and processing delays
- Keyword Management: Select/deselect keywords with bulk actions
- Professional UI: Two-column layout optimized for workflow efficiency
- Adobe Lightroom Classic 2024 (or newer)
- AI Provider (choose one):
- Google Gemini AI API Key (cloud-based, free tier available)
- Local Ollama installation (local processing, privacy-focused)
- Internet Connection (for Gemini AI only)
- Download the latest release from the plugin website
- Extract the ZIP file to a folder on your computer
- Open Adobe Lightroom Classic
- Go to File > Plug-in Manager
- Click Add button
- Navigate to the extracted plugin folder and select it
- Click Done
- Visit Google AI Studio
- Sign in with your Google account
- Click Get API Key
- Create a new API key or use an existing one
- Copy the API key (keep it secure!)
-
Install Ollama:
-
Install a vision model:
# Install the recommended LLaVA model ollama pull llava:latest # Or install a smaller/faster model ollama pull llava:7b
-
Start Ollama server:
ollama serve
The server will run on
http://localhost:11434by default. -
Verify installation:
ollama list
You should see your installed models listed.
- In Lightroom, go to File > Plug-in Manager
- Select AI Image Tagger from the list
- Choose your AI Provider:
- Google Gemini: Enter your API key in the API Key field
- Ollama (Local): Configure server URL and model name
- Configure your preferred settings:
- Choose which IPTC metadata fields to save
- Set batch processing options
- Customize AI prompts (optional)
- Click Done
- Select one or more photos in Lightroom's Library module
- Go to Library > Plug-in Extras > Tag Photos with AI
- Wait for AI analysis to complete (speed depends on your chosen provider)
- Review and edit the generated metadata:
- Caption: Brief description
- Description: Detailed description
- Keywords: Select relevant keywords
- Instructions: Usage suggestions
- Copyright: Attribution info
- Location: Identified landmarks
- Click Apply for current photo or Apply All for all photos
- Enable Use custom prompt in plugin settings
- Write your own AI analysis instructions
- Tailor the AI output to your specific needs
- Configure Batch Size (1-10 photos per batch)
- Set Delay Between Requests to avoid rate limiting
- Monitor progress in the analysis dialog
- Click Export Results to save analysis data as CSV
- Files saved as
aiimagetagger_results_YYYYMMDD_HHMMSS.csv - Includes all metadata fields and processing times
- Perfect for workflow documentation and analysis
| Feature | Google Gemini | Ollama (Local) |
|---|---|---|
| Cost | Free tier: 1,500 requests/day | Completely free |
| Privacy | Images sent to Google | All processing local |
| Speed | Fast (cloud processing) | Depends on hardware |
| Internet Required | Yes | No |
| Setup Complexity | Simple (API key only) | Moderate (software install) |
| Model Quality | High (latest Gemini models) | Good (LLaVA family) |
| Hardware Requirements | None | 8GB+ RAM recommended |
| Customization | Limited | Full control |
Choose Gemini if you want:
- Quick setup with just an API key
- Fast processing without local hardware requirements
- Latest AI capabilities with minimal configuration
Choose Ollama if you want:
- Complete privacy (no data leaves your computer)
- No ongoing costs or API limits
- Offline capability
- Full control over models and processing
- ✅ Save caption to IPTC metadata
- ✅ Save description to IPTC metadata
- ⬜ Save keywords to IPTC metadata (Note: Keywords automatically included in IPTC on export)
- ⬜ Save instructions to IPTC metadata
- ⬜ Save copyright to IPTC metadata
- ⬜ Save location to IPTC metadata
- Use custom prompt: Enable/disable custom AI instructions
- Custom Prompt: Write your own analysis instructions
- Batch Size: 1-10 photos (default: 5)
- Delay Between Requests: 500-5000ms (default: 1000ms)
- Create Keywords: Choose decoration style (as-is, prefix, suffix, parent)
- Keyword Value: Custom text for decoration
- API Key Security: Keys are stored securely in Lightroom's password storage
- Data Processing: Images are sent to Google's Gemini AI service for analysis
- No Storage: Google does not store your images after processing
- Local Metadata: All metadata management happens locally in Lightroom
- Complete Privacy: All processing happens on your local machine
- No Network Access: Images never leave your computer
- No API Keys: No external credentials required
- Local Storage: All models and data stored locally
"API key not configured"
- Ensure you've entered a valid Gemini AI API key in plugin settings
- Check that the API key has proper permissions
- Verify the key is active at Google AI Studio
"Network error"
- Verify your internet connection
- Check if your firewall allows Lightroom to access the internet
- Try increasing the delay between requests
"Ollama server not accessible"
- Ensure Ollama is running:
ollama serve - Check the server URL in plugin settings (default:
http://localhost:11434) - Verify Ollama is installed correctly:
ollama list
"Model not found"
- Install a vision model:
ollama pull llava:latest - Check available models:
ollama list - Verify the model name in plugin settings matches installed models
"Ollama connection timeout"
- Increase timeout in plugin settings (especially for larger models)
- Ensure sufficient RAM is available (8GB+ recommended)
- Consider using a smaller model like
llava:7b
"Analysis failed"
- Some images may not be suitable for AI analysis
- Try with different image formats (JPEG works best)
- Check the Lightroom log for detailed error messages
- Switch to the other AI provider to test
- Contact support: [email protected]
- 15 requests per minute
- 1,500 requests per day
- 1 million tokens per month
- Use batch processing with appropriate delays
- Process images during off-peak hours
- Consider upgrading to paid tier for heavy usage
We welcome contributions! Please see our Contributing Guide for details.
This project is licensed under the MIT License - see the LICENSE file for details.
- Created By: Anand's Photography (2025)
- Powered by: Google Gemini AI
- Built for: Adobe Lightroom Classic
Special thanks to @tjotala for lr-robotagger, which provided the inspiration to get started with this project.
Made with ❤️ for photographers and content creators worldwide