Releases · mmngays/llama.cpp

16 Jul 10:06

b5907 Latest

Latest

scripts: synthetic prompt mode for server-bench.py (#14695)

Assets 15

cudart-llama-bin-win-cuda-12.4-x64.zip

sha256:8c79a9b226de4b3cacfd1f83d24f962d0773be79f1e7b75c6af4ded7e32ae1d6

373 MB 2025-07-16T10:06:59Z
llama-b5907-bin-macos-arm64.zip

sha256:35614e62822b6b8cd556e0f621f4d87890e0b820352d0530024bd36bc242b5fd

10.6 MB 2025-07-16T10:07:09Z
llama-b5907-bin-macos-x64.zip

sha256:495e162cca4caa5561b33fcf7cede439daeeebade997ca86890d009cdfb5d171

26.4 MB 2025-07-16T10:07:10Z
llama-b5907-bin-ubuntu-vulkan-x64.zip

sha256:b65814cb0b802673b911263912315f274c8a43904783f7b49ba9da24a620e996

20.8 MB 2025-07-16T10:07:11Z
llama-b5907-bin-ubuntu-x64.zip

sha256:83cd4b910ef230df77dfbc375c51f90e5f7bf0d77803a5715363c17bdbf5c75a

12.4 MB 2025-07-16T10:07:13Z
llama-b5907-bin-win-cpu-arm64.zip

sha256:7281697a7c7ba315c7319572ea8e8173932d624e906626eebdafd51c5a55e3d4

10.8 MB 2025-07-16T10:07:14Z
llama-b5907-bin-win-cpu-x64.zip

sha256:4c4a2557112b89ee65d59268037c82f8ddb07fd12d679145810f9ae995afd25f

13.7 MB 2025-07-16T10:07:15Z
llama-b5907-bin-win-cuda-12.4-x64.zip

sha256:a05bc316335220c256b958c731c31319a4184c0e83e9d9624a558e7dcec36f8f

129 MB 2025-07-16T10:07:16Z
llama-b5907-bin-win-hip-radeon-x64.zip

sha256:9b7377d1bdfb0545ed995bc257894c86334784cabd3b4ada13c29d44b0b42670

298 MB 2025-07-16T10:07:20Z
llama-b5907-bin-win-opencl-adreno-arm64.zip

sha256:422d35ff3463e1be2ed34740be8cbdc75e26f0fcca994259683eb2fe8f951f08

11.2 MB 2025-07-16T10:07:29Z
Source code (zip)

2025-07-16T07:33:28Z
Source code (tar.gz)

2025-07-16T07:33:28Z

02 May 17:23

b5265

llama-model : support Qwen2 embedding models and pooling_mode_lasttok…

Assets 26

Provide feedback