Raw Metadata XML
<pkgmetadata>
<maintainer type="person">
<email>lucascs@proton.me</email>
<name>Lucas C.S.</name>
</maintainer>
<longdescription lang="en">
llama.cpp is an inference engine for large language models written in plain
C/C++, with no external runtime dependencies. It loads quantized models in
the GGUF format and runs them on CPU, or with optional acceleration through
CUDA, ROCm/HIP, Vulkan or OpenCL backends.
The package provides llama-cli (interactive and one-shot inference),
llama-server (an HTTP server exposing an OpenAI-compatible API, optionally
with an embedded WebUI), llama-quantize, llama-bench and the ggml libraries
used by many downstream projects.
Models are not shipped with this package and must be downloaded separately,
typically from Hugging Face. The CPU backend is selected through
CPU_FLAGS_X86, so enabling the AVX-512 and AMX flags supported by the host
processor has a large impact on inference throughput.
Note that ebuild is ported from the gentoo-zh overlay (sci-ml/llama-cpp).
</longdescription>
<use>
<flag name="blis">Build a BLIS backend</flag>
<flag name="flexiblas">Build a FlexiBLAS backend</flag>
<flag name="openblas">Build an OpenBLAS backend</flag>
<flag name="rocm">Build a HIP (ROCm) backend</flag>
<flag name="wmma">Use rocWMMA to enhance flash attention performance</flag>
<flag name="opencl">Build an OpenCL backend, so far only works on Adreno and Intel GPUs</flag>
<flag name="rpc">Build with rpc-server</flag>
<flag name="server">Build with example server</flag>
<flag name="webui">Build server with embedded WebUI</flag>
</use>
<upstream>
<changelog>https://github.com/ggml-org/llama.cpp/releases</changelog>
<bugs-to>https://github.com/ggml-org/llama.cpp/issues</bugs-to>
<remote-id type="github">ggml-org/llama.cpp</remote-id>
</upstream>
</pkgmetadata>