Install this version:
emerge -a =dev-python/amd-quark-bin-0.11.2
If this version is masked, you can unmask it using the autounmask tool or standard emerge options:
autounmask =dev-python/amd-quark-bin-0.11.2
Or alternatively:
emerge --autounmask-write -a =dev-python/amd-quark-bin-0.11.2
| Version | EAPI | Keywords | Slot |
|---|---|---|---|
| 0.11.2 | 8 | -* ~amd64 | 0 |
# Copyright 1999-2026 Gentoo Authors
# Distributed under the terms of the GNU General Public License v2
EAPI=8
DISTUTILS_USE_PEP517=no
PYTHON_COMPAT=( python3_{11,12} )
inherit distutils-r1
MY_PN=${PN%-bin}
MY_WHEEL="${MY_PN//-/_}-${PV}-py3-none-any.whl"
DESCRIPTION="AMD's quantization toolkit for ML model optimization (binary wheel)"
HOMEPAGE="
https://quark.docs.amd.com
https://github.com/amd/quark
https://pypi.org/project/amd-quark/
"
SRC_URI="
https://files.pythonhosted.org/packages/95/59/9191965320615f58d582eb805661484f741dac97d6e10c7989f051a6aa7e/${MY_WHEEL}
"
S="${WORKDIR}"
LICENSE="MIT"
SLOT="0"
KEYWORDS="-* ~amd64"
# Wheel-only on PyPI; upstream pins requires_python = "<3.13,>=3.9".
# Required by dev-python/vllm in the rocm path.
#
# Two upstream caps relaxed because ::gentoo doesn't carry the older
# versions:
# - numpy<=2.1.3 (upstream) → ::gentoo only has >=2.2.6. Likely
# conservative; runtime issues on numpy >=2.2 if hit will be
# visible as AttributeError on internal code paths.
# - onnx<=1.19.0 (upstream) → ::gentoo has 1.18.0 + 1.20.1; pinned
# to 1.18.x to stay below the upstream cap.
#
# onnxscript and onnxslim (declared as install-time deps of the wheel)
# are unpackaged in ::gentoo + ::guru — runtime paths that touch ONNX
# export will fail with ImportError, but the PyTorch-only quantization
# paths vllm normally invokes don't need them.
RDEPEND="
sci-ml/evaluate[${PYTHON_USEDEP}]
dev-python/joblib[${PYTHON_USEDEP}]
app-alternatives/ninja
dev-python/numpy[${PYTHON_USEDEP}]
~sci-ml/onnx-1.18.0[${PYTHON_USEDEP}]
dev-python/pandas[${PYTHON_USEDEP}]
dev-python/plotly[${PYTHON_USEDEP}]
dev-python/protobuf[${PYTHON_USEDEP}]
dev-python/pydantic[${PYTHON_USEDEP}]
dev-python/rich[${PYTHON_USEDEP}]
dev-python/scipy[${PYTHON_USEDEP}]
sci-ml/sentencepiece[${PYTHON_USEDEP}]
dev-python/tqdm[${PYTHON_USEDEP}]
dev-python/zstandard[${PYTHON_USEDEP}]
"
src_unpack() {
mkdir -p "${S}/wheel" || die
cp "${DISTDIR}/${MY_WHEEL}" "${S}/wheel/" || die
}
src_install() {
python_foreach_impl install_wheel
}
install_wheel() {
${EPYTHON} -m installer --destdir="${D}" "${S}/wheel/${MY_WHEEL}" || die
}
sci-ml/evaluate[] dev-python/joblib[] app-alternatives/ninja dev-python/numpy[] ~sci-ml/onnx-1.18.0[] dev-python/pandas[] dev-python/plotly[] dev-python/protobuf[] dev-python/pydantic[] dev-python/rich[] dev-python/scipy[] sci-ml/sentencepiece[] dev-python/tqdm[] dev-python/zstandard[]
| Type | File | Size | Source URLs |
|---|---|---|---|
| DIST | amd_quark-0.11.2-py3-none-any.whl | 1907249 bytes | https://files.pythonhosted.org/packages/95/59/9191965320615f58d582eb805661484f741dac97d6e10c7989f051a6aa7e/amd_quark-0.11.2-py3-none-any.whl |