Download Latest Version b7378 source code.tar.gz (28.1 MB)
Email in envelope

Get an email when there's a new version of llama.cpp

Home / b7372
Name Modified Size InfoDownloads / Week
Parent folder
llama-b7372-xcframework.zip < 19 hours ago 147.1 MB
llama-b7372-xcframework.tar.gz < 19 hours ago 147.2 MB
llama-b7372-bin-win-vulkan-x64.zip < 19 hours ago 33.6 MB
llama-b7372-bin-win-sycl-x64.zip < 19 hours ago 107.8 MB
llama-b7372-bin-win-opencl-adreno-arm64.zip < 19 hours ago 15.6 MB
llama-b7372-bin-win-hip-radeon-x64.zip < 19 hours ago 346.5 MB
llama-b7372-bin-win-cuda-13.1-x64.zip < 19 hours ago 91.3 MB
llama-b7372-bin-win-cuda-12.4-x64.zip < 19 hours ago 202.6 MB
llama-b7372-bin-win-cpu-x64.zip < 19 hours ago 18.7 MB
llama-b7372-bin-win-cpu-arm64.zip < 19 hours ago 15.1 MB
llama-b7372-bin-ubuntu-x64.zip < 19 hours ago 17.7 MB
llama-b7372-bin-ubuntu-x64.tar.gz < 19 hours ago 17.7 MB
llama-b7372-bin-ubuntu-vulkan-x64.zip < 19 hours ago 33.2 MB
llama-b7372-bin-ubuntu-vulkan-x64.tar.gz < 19 hours ago 33.2 MB
llama-b7372-bin-ubuntu-s390x.zip < 19 hours ago 17.6 MB
llama-b7372-bin-ubuntu-s390x.tar.gz < 19 hours ago 20.6 MB
llama-b7372-bin-macos-x64.zip < 19 hours ago 40.0 MB
llama-b7372-bin-macos-x64.tar.gz < 19 hours ago 40.0 MB
llama-b7372-bin-macos-arm64.zip < 19 hours ago 15.3 MB
llama-b7372-bin-macos-arm64.tar.gz < 19 hours ago 15.3 MB
cudart-llama-bin-win-cuda-13.1-x64.zip < 19 hours ago 402.6 MB
cudart-llama-bin-win-cuda-12.4-x64.zip < 19 hours ago 391.4 MB
b7372 source code.tar.gz 2025-12-12 28.1 MB
b7372 source code.zip 2025-12-12 29.0 MB
README.md 2025-12-12 1.9 kB
Totals: 25 Items   2.2 GB 0

[!WARNING] Release Format Update: Linux releases will soon use .tar.gz archives instead of .zip. Please make the necessary changes to your deployment scripts.

CUDA: fix overflow in MMA kernel without stream-k (#17939)

macOS/iOS: - macOS Apple Silicon (arm64) - macOS Intel (x64) - iOS XCFramework

Linux: - Ubuntu x64 (CPU) - Ubuntu x64 (Vulkan) - Ubuntu s390x (CPU)

Windows: - Windows x64 (CPU) - Windows arm64 (CPU) - Windows x64 (CUDA 12) - Windows x64 (CUDA 13) - Windows x64 (Vulkan) - Windows x64 (SYCL) - Windows x64 (HIP)

Source: README.md, updated 2025-12-12