marcsun13
/

gguf-kernels

Model card Files Files and versions

217 MB

Ctrl+K

Ctrl+K

1 contributor

History: 12 commits

marcsun13's picture

marcsun13 HF Staff

Minimal README: ops, devices, source and how to refresh it

fd5d3aa 37 minutes ago

build
Refresh the Metal builds after the command-buffer fix about 1 hour ago
gguf_cuda
GGUF kernels: dequantize + fused gemv over packed blocks, 9 CUDA variants 2 days ago
gguf_metal
Encode on the MPS stream's own queue about 2 hours ago
tests
GGUF kernels: dequantize + fused gemv over packed blocks, 9 CUDA variants 2 days ago
torch-ext
GGUF kernels: dequantize + fused gemv over packed blocks, 9 CUDA variants 2 days ago
vendor
Add Metal backend; vendor whole llama.cpp trees about 22 hours ago
.gitattributes

3.63 kB
Refresh the Metal builds after the command-buffer fix about 1 hour ago
.gitignore

95 Bytes
Minimal README: ops, devices, source and how to refresh it 37 minutes ago
README.md

1.69 kB
Minimal README: ops, devices, source and how to refresh it 37 minutes ago
SKILL.md

8.62 kB
GGUF kernels: dequantize + fused gemv over packed blocks 2 days ago
build.toml

3.43 kB
Add Metal backend; vendor whole llama.cpp trees about 22 hours ago
flake.lock

3.05 kB
Add Metal backend; vendor whole llama.cpp trees about 22 hours ago
flake.nix

290 Bytes
GGUF kernels: dequantize + fused gemv over packed blocks 2 days ago
vendor.py

3.29 kB
Add Metal backend; vendor whole llama.cpp trees about 22 hours ago