b8699
📦 llama-cppView on GitHub →
✨ 1 features🐛 1 fixes🔧 1 symbols
Summary
This release introduces support for attention rotation within the kv-cache for heterogeneous iSWA configurations and removes an unnecessary assertion.
✨ New Features
- kv-cache now supports attention rotation for heterogeneous iSWA.
🐛 Bug Fixes
- Removed an assertion in the continuous integration (cont) checks.