sci-ml/llama-cpp-python (local-ai)

Search

Install

Install this package:

emerge -a sci-ml/llama-cpp-python

Package Information

Description:
Python bindings for the llama.cpp library
Homepage:
https://github.com/abetlen/llama-cpp-python https://pypi.org/project/llama-cpp-python/
License:
MIT

Versions

Version EAPI Keywords Slot
0.3.36 8 ~amd64 0

Metadata

Maintainers

Upstream

Raw Metadata XML
<pkgmetadata>
	<maintainer type="person">
		<email>plamen@ipnmod.org</email>
		<name>Plamen K. Kosseff</name>
	</maintainer>
	<use>
		<flag name="native">Let ggml tune the CPU kernels for the build host (-march=native) instead of the CPU_FLAGS_X86 gates</flag>
		<flag name="openblas">Route matrix multiplication through <pkg>sci-libs/openblas</pkg></flag>
		<flag name="vulkan">Build the ggml Vulkan backend for GPU offload</flag>
	</use>
	<upstream>
		<remote-id type="github">abetlen/llama-cpp-python</remote-id>
		<remote-id type="pypi">llama-cpp-python</remote-id>
	</upstream>
</pkgmetadata>

Lint Warnings

USE Flags

Manage flags for this package: euse -i <flag> -p sci-ml/llama-cpp-python | euse -E <flag> -p sci-ml/llama-cpp-python | euse -D <flag> -p sci-ml/llama-cpp-python

Flag Description 0.3.36
( ⚠️ ✓
) ⚠️ ✓
amx_bf16 ⚠️ ✓
amx_int8 ⚠️ ✓
amx_tile ⚠️ ✓
avx ⚠️ ✓
avx2 ⚠️ ✓
avx512_bf16 ⚠️ ✓
avx512_vnni ⚠️ ✓
avx512bw ⚠️ ✓
avx512f ⚠️ ✓
avx512vbmi ⚠️ ✓
avx_vnni ⚠️ ✓
bmi2 ⚠️ ✓
f16c ⚠️ ✓
fma3 ⚠️ ✓
native Let ggml tune the CPU kernels for the build host (-march=native) instead of the CPU_FLAGS_X86 gates ✓
openblas Route matrix multiplication through <pkg>sci-libs/openblas</pkg> ✓
openmp ⚠️ ⊕
sse4_2 ⚠️ ✓
video_cards_amdgpu ⚠️ ✓
vulkan Build the ggml Vulkan backend for GPU offload ✓

Manifest

Type File Size Versions
Unmatched Entries
Type File Size
DIST llama_cpp_python-0.3.36.tar.gz 76589250 bytes
EBUILD llama-cpp-python-0.3.36.ebuild 3149 bytes
MISC metadata.xml 709 bytes