b8364
📦 llama-cppView on GitHub →
🐛 1 fixes
Summary
This release primarily focuses on platform-specific binary updates and includes a fix to limit the number of FA stream-k CUDA blocks.
🐛 Bug Fixes
- Limited the number of FA stream-k CUDA blocks.
This release primarily focuses on platform-specific binary updates and includes a fix to limit the number of FA stream-k CUDA blocks.