b8180
📦 llama-cppView on GitHub →
✨ 3 features🐛 3 fixes🔧 2 symbols
Summary
This release introduces model metadata loading from Hugging Face for testing purposes, along with optimizations for incremental downloading and fixes related to compilation conditions.
✨ New Features
- Added model metadata loading from Hugging Face for use in tests.
- Implemented incremental chunking for model downloads instead of full redownloads.
- Added support for loading metadata from individual split files when dealing with split models, and avoiding mmproj loading.
🐛 Bug Fixes
- Fixed caching issue related to model metadata loading.
- Added a warning when incremental model metadata loading fails.
- Fixed compilation only occurring when cpp-httplib has SSL support.