Install this version:
emerge -a =dev-python/quack-kernels-0.3.3If this version is masked, you can unmask it using the autounmask tool or standard emerge options:
autounmask =dev-python/quack-kernels-0.3.3Or alternatively:
emerge --autounmask-write -a =dev-python/quack-kernels-0.3.3| Version | EAPI | Keywords | Slot |
|---|---|---|---|
| 0.3.3 | 8 | ~amd64 | 0 |
# Copyright 1999-2026 Gentoo Authors
# Distributed under the terms of the GNU General Public License v2
EAPI=8
DISTUTILS_USE_PEP517=setuptools
PYTHON_COMPAT=( python3_{11..14} )
inherit distutils-r1 pypi
DESCRIPTION="QuACK attention kernels for vLLM/FlashInfer (cute-DSL based)"
HOMEPAGE="
https://github.com/Dao-AILab/quack
https://pypi.org/project/quack-kernels/
"
LICENSE="Apache-2.0"
SLOT="0"
KEYWORDS="~amd64"
# Pure-Python; the kernels themselves are JIT-compiled at first use
# via apache-tvm-ffi + nvidia-cutlass-dsl. Upstream's cu13 extra
# selects the matching cutlass-dsl[cu13] sub-libs which we already
# pull via dev-python/nvidia-cutlass-dsl on this overlay; no USE
# flag needed. # verified 2026-05-07 against 0.3.3.
RDEPEND="
>=dev-python/nvidia-cutlass-dsl-4.4.1[${PYTHON_USEDEP}]
>=dev-python/apache-tvm-ffi-0.1.6[${PYTHON_USEDEP}]
<dev-python/apache-tvm-ffi-0.2[${PYTHON_USEDEP}]
dev-python/torch-c-dlpack-ext[${PYTHON_USEDEP}]
dev-python/einops[${PYTHON_USEDEP}]
sci-ml/pytorch[${PYTHON_USEDEP}]
"
DEPEND="${RDEPEND}"
>=dev-python/nvidia-cutlass-dsl-4.4.1[] >=dev-python/apache-tvm-ffi-0.1.6[] <dev-python/apache-tvm-ffi-0.2[] dev-python/torch-c-dlpack-ext[] dev-python/einops[] sci-ml/pytorch[]
>=dev-python/nvidia-cutlass-dsl-4.4.1[] >=dev-python/apache-tvm-ffi-0.1.6[] <dev-python/apache-tvm-ffi-0.2[] dev-python/torch-c-dlpack-ext[] dev-python/einops[] sci-ml/pytorch[]