Change8

b8944

📦 llama-cppView on GitHub →
1 features🔧 1 symbols

Summary

This release updates ggml to use 64 bytes aligned tile buffers, showing minor speedups across various Qwen 35 quantization tests. It also provides extensive pre-compiled binaries for numerous operating systems and hardware configurations.

✨ New Features

  • ggml: Implemented 64 bytes aligned tile buffers for potential performance improvements.

Affected Symbols