b8070
📦 llama-cppView on GitHub →
✨ 2 features🐛 1 fixes🔧 1 symbols
Summary
This release introduces model graph deduplication and updates for Qwen family models, including the addition of `llm_build_delta_net_base`, alongside providing numerous pre-compiled binaries for diverse platforms.
✨ New Features
- Added support for deduplicating delta-net graphs for the Qwen family of models.
- Introduced the new build target `llm_build_delta_net_base` for models.
🐛 Bug Fixes
- Ensured that Qwen35 and Qwen35moe graphs remain intact during processing.