Releases · MoonRide303/llama.cpp · GitHub

28 May 16:34

b5523 Latest

Latest

convert: small addition to support LlamaModel (#13838)

Co-authored-by: dinhhuy <[email protected]>

Assets 18

cudart-llama-bin-win-cuda-11.7-x64.zip

303 MB 2025-05-28T16:34:01Z
cudart-llama-bin-win-cuda-12.4-x64.zip

373 MB 2025-05-28T16:34:10Z
llama-b5523-bin-macos-arm64.zip

10.9 MB 2025-05-28T16:34:26Z
llama-b5523-bin-macos-x64.zip

26.1 MB 2025-05-28T16:34:27Z
llama-b5523-bin-ubuntu-arm64.zip

11.8 MB 2025-05-28T16:34:28Z
llama-b5523-bin-ubuntu-vulkan-x64.zip

20.1 MB 2025-05-28T16:34:29Z
llama-b5523-bin-ubuntu-x64.zip

12.3 MB 2025-05-28T16:34:30Z
llama-b5523-bin-win-cpu-arm64.zip

11.1 MB 2025-05-28T16:34:31Z
llama-b5523-bin-win-cpu-x64.zip

13.7 MB 2025-05-28T16:34:31Z
llama-b5523-bin-win-cuda-11.7-x64.zip

109 MB 2025-05-28T16:34:32Z
Source code (zip)

2025-05-28T14:34:18Z
Source code (tar.gz)

2025-05-28T14:34:18Z

06 May 08:02

b5289

CUDA: fix --split-mode row for MMQ (#13323)

Assets 21