Search Criteria
Package Details: llama.cpp-sycl 2026.0.0-2
Package Actions
| Git Clone URL: | https://aur.archlinux.org/llama.cpp-sycl.git (read-only, click to copy) |
|---|---|
| Package Base: | llama.cpp-sycl |
| Description: | llama.cpp with Intel Arc GPU acceleration via SYCL/oneAPI (bundled runtime, no separate oneAPI install required). Please read the README on GitHub before use. |
| Upstream URL: | https://github.com/cantosun99/llama.cpp-sycl |
| Keywords: | arc cpp intel llama llama.cpp oneapi sycl vulkan |
| Licenses: | MIT |
| Conflicts: | intel-oneapi-toolkit, llama.cpp, llama.cpp-bin, llama.cpp-bin-noavx, llama.cpp-clblas-git, llama.cpp-clblast, llama.cpp-cublas-git, llama.cpp-cuda, llama.cpp-cuda-essentials-only, llama.cpp-cuda-git, llama.cpp-gfx1151, llama.cpp-git, llama.cpp-hip, llama.cpp-hip-gfx1151, llama.cpp-hipblas-git, llama.cpp-opencl, llama.cpp-openvino, llama.cpp-sycl-f16, llama.cpp-sycl-f16-git, llama.cpp-sycl-f32, llama.cpp-sycl-f32-git, llama.cpp-vulkan, llama.cpp-vulkan-bin, llama.cpp-vulkan-gemma4, llama.cpp-vulkan-git |
| Provides: | llama.cpp |
| Submitter: | cantosun99 |
| Maintainer: | cantosun99 |
| Last Packager: | cantosun99 |
| Votes: | 0 |
| Popularity: | 0.000000 |
| First Submitted: | 2026-05-06 09:00 (UTC) |
| Last Updated: | 2026-05-13 08:50 (UTC) |
Dependencies (7)
- gcc-libs (gcc-libs-gitAUR, gccrs-libs-gitAUR, gcc-libs-snapshotAUR)
- intel-compute-runtime (intel-compute-runtime-legacy-binAUR, intel-compute-runtime-gitAUR, intel-compute-runtime-legacyAUR, intel-compute-runtime-binAUR)
- level-zero-headers (level-zero-headers-gitAUR, level-zero-headers-legacyAUR)
- level-zero-loader (level-zero-loader-gitAUR, level-zero-loader-legacyAUR)
- cmake (cmake3AUR, cmake-gitAUR) (make)
- git (git-gitAUR, git-glAUR, git-wd40AUR) (make)
- make (make-gitAUR, make-staticAUR) (make)
Required by (5)
- llamaman-bin (requires llama.cpp) (optional)
- scmd-bin (requires llama.cpp)
- voxd (requires llama.cpp) (optional)
- voxd-bin (requires llama.cpp) (optional)
- voxd-git (requires llama.cpp) (optional)
Latest Comments
cantosun99 commented on 2026-05-06 09:02 (UTC)
This package exists because getting SYCL acceleration for Intel Arc GPUs on Arch is genuinely painful. The full intel-oneapi-toolkit is gigabytes of stuff you don't need, and the existing llama.cpp-sycl-fp16/fp32 packages leave the oneAPI setup to you. This package bundles only the essential oneAPI components and builds llama.cpp from source on your machine, so you install one package and you are done. If you have an Intel Arc GPU, you want SYCL over Vulkan. Arc's XMX units (Intel's equivalent of tensor cores) are only exposed through SYCL/Level Zero, not through Vulkan compute. Intel also actively develops and tests llama.cpp's SYCL backend, whereas Vulkan on Intel is more of a community-maintained path. The performance difference in LLM inference is real. Tested on a Sparkle Intel Arc Pro B70 passive running Unsloth's Qwen3 27B Q6_K on CachyOS with the Linux 7.0.3-1-cachyos kernel. First AUR package, feedback welcome. Contact: privat@cantosun.de