Install this package:
emerge -a sci-misc/llama-cpp
If the package is masked, you can unmask it using the autounmask tool or standard emerge options:
autounmask sci-misc/llama-cpp
Or alternatively:
emerge --autounmask-write -a sci-misc/llama-cpp
<pkgmetadata> <maintainer type="person"> <email>vklimovs@gmail.com</email> <name>Vjaceslavs Klimovs</name> </maintainer> <use> <flag name="blis">Build a BLIS BLAS backend</flag> <flag name="flexiblas">Build a FlexiBLAS BLAS backend</flag> <flag name="openblas">Build an OpenBLAS BLAS backend</flag> <flag name="opencl">Build an OpenCL backend (Adreno and Intel GPUs only)</flag> <flag name="openssl">Use OpenSSL to support HTTPS in the server</flag> <flag name="rocm">Build a HIP (ROCm) backend for AMD GPUs</flag> <flag name="wmma">Use rocWMMA to accelerate Flash Attention on supported AMD GPUs</flag> </use> <upstream> <remote-id type="github">ggml-org/llama.cpp</remote-id> </upstream> </pkgmetadata>
Manage flags for this package:
euse -i <flag> -p sci-misc/llama-cpp |
euse -E <flag> -p sci-misc/llama-cpp |
euse -D <flag> -p sci-misc/llama-cpp
| Flag | Description | 9999 | 0_pre9082 |
|---|---|---|---|
| blis | Build a BLIS BLAS backend | ✓ | ✓ |
| cuda | ⚠️ | ✓ | ✓ |
| curl | Enable curl support for HTTP transfers in scripts ⚠️ | ✓ | ✓ |
| flexiblas | Build a FlexiBLAS BLAS backend | ✓ | ✓ |
| openblas | Build an OpenBLAS BLAS backend | ✓ | ✓ |
| opencl | Build an OpenCL backend (Adreno and Intel GPUs only) | ✓ | ✓ |
| openmp | ⚠️ | ⊕ | ⊕ |
| openssl | Use OpenSSL to support HTTPS in the server | ✓ | ✓ |
| rocm | Build a HIP (ROCm) backend for AMD GPUs | ✓ | ✓ |
| vulkan | ⚠️ | ✓ | ✓ |
| wmma | Use rocWMMA to accelerate Flash Attention on supported AMD GPUs | ✓ | ✓ |
| Type | File | Size | Versions |
|---|---|---|---|
| DIST | llama-cpp-0_pre9082.tar.gz | 34031348 bytes | 0_pre9082 |
| Type | File | Size |
|---|