Releases · thevishalagarwal/llama.cpp · GitHub

12 May 12:17

b5355 Latest

Latest

llama-bench : accept ranges for integer parameters (#13410)

Assets 20

cudart-llama-bin-win-cuda11.7-x64.zip

303 MB 2025-05-12T12:17:17Z
cudart-llama-bin-win-cuda12.4-x64.zip

373 MB 2025-05-12T12:17:24Z
llama-b5355-bin-macos-arm64.zip

9.86 MB 2025-05-12T12:17:34Z
llama-b5355-bin-macos-x64.zip

23.4 MB 2025-05-12T12:17:35Z
llama-b5355-bin-ubuntu-arm64.zip

10.5 MB 2025-05-12T12:17:36Z
llama-b5355-bin-ubuntu-vulkan-x64.zip

18.6 MB 2025-05-12T12:17:37Z
llama-b5355-bin-ubuntu-x64.zip

11.1 MB 2025-05-12T12:17:38Z
llama-b5355-bin-win-cpu-arm64.zip

11.5 MB 2025-05-12T12:17:39Z
llama-b5355-bin-win-cpu-x64.zip

12.6 MB 2025-05-12T12:17:40Z
llama-b5355-bin-win-cuda11.7-x64.zip

127 MB 2025-05-12T12:17:41Z
Source code (zip)

2025-05-12T11:08:22Z
Source code (tar.gz)

2025-05-12T11:08:22Z

07 May 17:08

github-workflow-update-cuda-12.8-b5305-6bcceca

github-workflow-update-cuda-12.8-b5305-6bcceca

update cuda to 12.8 for CI

Assets 19