b9568
📦 llama-cppView on GitHub →
✨ 3 features🐛 2 fixes🔧 3 symbols
Summary
This release introduces MTP support for gemma-4 E2B and E4B assistants by updating model conversion logic and architecture support. It also provides numerous pre-compiled binaries across different operating systems and hardware configurations.
✨ New Features
- Added support for gemma-4 E2B and E4B assistants via MTP.
- Updated model converter to support smaller assistants.
- Added masked_embd tensors to gemma4-assist architecture.
🐛 Bug Fixes
- Removed temporary debug code related to gemma-4 conversion.
- Filtered out masked_embedding tensors during gemma-4-mtp conversion.