b7621
📦 llama-cppView on GitHub →
🐛 1 fixes🔧 1 symbols
Summary
This release focuses on a bug fix for concurrent streams within the ggml-cuda implementation and provides updated binaries for multiple platforms including Windows, Linux, and macOS.
🐛 Bug Fixes
- ggml-cuda: fixes for concurrent streams (#18496)
🔧 Affected Symbols
ggml-cuda