b8951
📦 llama-cppView on GitHub →
✨ 1 features
Summary
This release introduces performance enhancements through new fast matrix-vector kernels optimized for i-quants. It also provides a comprehensive set of pre-compiled binaries for macOS, Linux, Android, Windows, and openEuler targeting various CPUs and accelerators.
✨ New Features
- Added fast matrix-vector kernels for i-quants.