b8606
📦 llama-cppView on GitHub →
✨ 1 features🔧 1 symbols
Summary
This release focuses on migrating ggml-webgpu operations from AOT to JIT compilation and provides numerous pre-compiled binaries across various platforms and hardware configurations.
Migration Steps
- If you were relying on AOT template expansion in ggml-webgpu, note that unused functions from embed_wgsl.py have been removed.
✨ New Features
- Ported all AOT operators in ggml-webgpu to use JIT compilation for cpy, glu, rope, and soft_max pipelines.