b9142
📦 llama-cppView on GitHub →
✨ 2 features🐛 2 fixes🔧 1 symbols
Summary
This release focuses on expanding OpenCL support for MoE models on Adreno GPUs, adding q5_0 and q5_1 quantization levels, alongside general stability improvements.
✨ New Features
- Added q5_0 MoE support for OpenCL on Adreno.
- Added q5_1 MoE support for OpenCL on Adreno.
🐛 Bug Fixes
- Avoided potential memory leak in OpenCL implementation.
- Suppressed unused variable warnings when building OpenCL for non-Adreno targets.