marcsun13
/

ggml-quantization

Model card Files Files and versions

ggml-quantization

477 kB

Ctrl+K

Ctrl+K

2 contributors

History: 36 commits

marcsun13's picture

marcsun13 HF Staff

correct the versioning note: a version is a branch on the kernel repo, not a tag

35c770e 3 days ago

gguf_metal
revert the f32 gemv: measured no end-to-end gain, and nothing dispatches it 3 days ago
tests
expose ggml's get_rows, and rename the package to match the repo 7 days ago
torch-ext
add mul_mat_id, and compile upstream's kernels as they ship 6 days ago
vendor
add mul_mat_id, and compile upstream's kernels as they ship 6 days ago
.gitattributes

1.76 kB
rebuild 6 days ago
.gitignore

95 Bytes
Minimal README: ops, devices, source and how to refresh it 27 days ago
README.md

1.36 kB
add mul_mat_id, and compile upstream's kernels as they ship 6 days ago
SKILL.md

14.4 kB
correct the versioning note: a version is a branch on the kernel repo, not a tag 3 days ago
build.toml

1.79 kB
add mul_mat_id, and compile upstream's kernels as they ship 6 days ago
flake.lock

3.05 kB
Add Metal backend; vendor whole llama.cpp trees 28 days ago
flake.nix

294 Bytes
vendor only the files the build reaches 6 days ago
vendor.py

2.48 kB
add mul_mat_id, and compile upstream's kernels as they ship 6 days ago