b9855
📦 llama-cppView on GitHub →
✨ 2 features🔧 1 symbols
Summary
This release introduces performance improvements to ggml-cpu by adding AVX2 optimization for nvfp4 dot product and integrating UE4M3 LUTs. Various pre-compiled binaries for different platforms are provided.
✨ New Features
- Added AVX2 optimization for nvfp4 dot product in ggml-cpu.
- Added UE4M3 LUT usage for nvfp4 dot product in ggml-cpu.