model-registry-maintainer

Maintains MassGen's model registry by updating model details in capabilities.py and token_manager.py for new releases, pricing changes, or feature updates.

• Manages model capabilities, release dates, and features in capabilities.py
• Handles pricing and context limits in token_manager.py
• Ensures registry syncs with provider updates and new model releases
• Updates both files when new models, pricing, or features are added

• Updates model release dates, pricing, and context windows
• Uses LiteLLM database as primary pricing source (cached 1 hour)
• Fallback pricing via hardcoded PROVIDER_PRICING for models not in LiteLLM
• Maintains accurate model capabilities (e.g., web search, vision)
• Validates changes via tests and config generation
• Generates updated documentation after updates

1. 1.Add new model to capabilities.py with release date and features
2. 2.Check LiteLLM database for existing pricing; add to PROVIDER_PRICING only if missing
3. 3.Update capabilities or default model if new features or defaults are introduced
4. 4.Run tests and config generation to validate changes
5. 5.Regenerate backend documentation after updates