b8880
📦 llama-cppView on GitHub →
🐛 2 fixes🔧 1 symbols
Summary
This release focuses on internal improvements for ggml-webgpu, specifically resetting profiling times upon context freeing and restructuring GPU profiling context.
🐛 Bug Fixes
- Reset the CPU/GPU profiling time when freeing context in ggml-webgpu.
- Moved GPU profiling time from global context to webgpu_context in ggml-webgpu.