ik_llama-cpp
llama.cpp fork with additional SOTA quants and improved performance
homepage ↗ github: ikawrakow/ik_llama.cpp
Available in
| Overlay | Newest | Ebuilds | Last activity | |
|---|---|---|---|---|
| guru gitweb ↗ | 9999 | 1 | 46 h | details › |
Versions & arches
Use flags of 9999
- curl Add support for client-side URL transfer library
- openblas Build an OpenBLAS backend
- +openmp Build support for the OpenMP (support parallel computing), requires >=sys-devel/gcc-4.2 built with USE="openmp"
- blis Build a BLIS backend
- rocm Build a HIP (ROCm) backend
- cuda Enable NVIDIA CUDA support (computation on GPU)
- vulkan Add support for 3D graphics and computing via the Vulkan cross-platform API
- flexiblas Build a FlexiBLAS backend
- wmma Use rocWMMA to enhance flash attention performance
- +amdgpu_targets_gfx908
- +amdgpu_targets_gfx90a
- +amdgpu_targets_gfx942
- +amdgpu_targets_gfx1030
- +amdgpu_targets_gfx1100
- +amdgpu_targets_gfx1101
- +amdgpu_targets_gfx1200
- +amdgpu_targets_gfx1201
- amdgpu_targets_gfx803
- amdgpu_targets_gfx900
- amdgpu_targets_gfx906
- amdgpu_targets_gfx940
- amdgpu_targets_gfx941
- amdgpu_targets_gfx1010
- amdgpu_targets_gfx1011
- amdgpu_targets_gfx1012
- amdgpu_targets_gfx1031
- amdgpu_targets_gfx1102
- amdgpu_targets_gfx1103
- amdgpu_targets_gfx1150
- amdgpu_targets_gfx1151
Runtime dependencies of 9999
show 28 lines
curl?
(
)
openblas?
(
)
openmp?
(
)
blis?
(
)
flexiblas?
(
)
rocm?
(
wmma?
(
)
)
cuda?
(
)
vulkan?
(
)