Change8

b8192

📦 llama-cppView on GitHub →
1 features

Summary

This release introduces a new FP16 compute path for q4_0 GEMM on aarch64 systems and provides updated pre-compiled binaries across multiple platforms.

✨ New Features

  • Added an FP16 compute path for q4_0 GEMM operations on aarch64 architectures.