Releases: lemonade-sdk/llama.cpp
Releases · lemonade-sdk/llama.cpp
b9827: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9827-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9827-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9827-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9824: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9824-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9824-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9824-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9821: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9821-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9821-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9821-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9820: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9820-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9820-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9820-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9803: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9803-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9803-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9803-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9801: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9801-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9801-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9801-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9797: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9797-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9797-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9797-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9793: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9793-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9793-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9793-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9786: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9786-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9786-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9786-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)
b9781: Merge pull request #18 from lemonade-sdk/kenvandine/cuda_sm_121
Nightly release for ea05f50
Linux:
- Ubuntu x64 (ROCm 7.13)
- Ubuntu x64 (CUDA):
llama-b9781-ubuntu-cuda-sm_XX-x64.tar.xz(replace XX with your GPU compute capability) - Ubuntu arm64 (CUDA):
llama-b9781-ubuntu-cuda-sm_XX-arm64.tar.xz(replace XX with your GPU compute capability) - Ubuntu x64 (OpenVINO 2026.0)
Windows:
- Windows x64 (ROCm 7.13)
- Windows x64 (CUDA):
llama-b9781-windows-cuda-sm_XX-x64.7z(replace XX with your GPU compute capability)