euse -i cudaeuse -E cudaeuse -D cudaEnable NVIDIA CUDA support (computation on GPU)
| Package | Local Description (use.local.desc) | Metadata Description (metadata.xml) |
|---|---|---|
| sci-misc/llama-cpp | - | - |
| sci-libs/tensorflow | - | - |
| media-gfx/blender | Build cycles renderer with nVidia CUDA support. | Build cycles renderer with nVidia CUDA support. |
| sci-libs/ceres-solver | - | - |
| dev-python/comfy-kitchen | - | - |
| media-gfx/comfyui | - | Build the torch backend (<pkg>sci-ml/caffe2</pkg>) with CUDA/NVIDIA acceleration; also selects the CUDA wheels of <pkg>dev-python/comfy-aimdo-bin</pkg> and <pkg>dev-python/comfy-kitchen-bin</pkg> (otherwise their non-CUDA wheels) |
| sci-ml/colibri | - | Build with NVIDIA CUDA GPU offloading for resident expert tensors |
| dev-util/goose | - | - |
| media-gfx/blender-bin | - | - |
| media-gfx/openvdb | Enable support for CUDA in NanoVDB | Enable support for CUDA in NanoVDB |
| media-libs/opencv | Enable NVIDIA Cuda computations support (Experimental!) | Enable NVIDIA Cuda computations support (Experimental!) |
| sci-ml/ik_llama-cpp | - | - |
| sci-ml/koboldcpp | Build the CUDA/cuBLAS backend (koboldcpp_cublas) for NVIDIA GPUs, for every architecture the installed toolkit supports | Build the CUDA/cuBLAS backend (koboldcpp_cublas) for NVIDIA GPUs, for every architecture the installed toolkit supports |
| sci-ml/llama-cpp | - | ggml: use CUDA ; in order for it to compile also tag .... "cpu" has to be selected, don't ask why. Use CMAKE_EXTRA_CACHE_FILE env variable (check https://gitweb.gentoo.org/repo/gentoo.git/tree/eclass/cmake.eclass ) to specify next variables: CMAKE_CUDA_ARCHITECTURES ( use next command to find native architecture: `nvidia-smi --query-gpu=compute_cap --format=csv | tail -n 1 | sed -e 's/\.//g'` ) ; GGML_CUDA_PEER_MAX_BATCH_SIZE ggml: max. batch size for using peer access, default: 128 |
| sci-ml/ollama-bin | Enable NVIDIA CUDA GPU acceleration support. Requires compatible NVIDIA GPU (compute capability 6.0+, Pascal architecture or newer) and nvidia-cuda-toolkit. Significantly improves inference performance for large models. | Support for Nvidia CUDA compatible cards |
| sci-ml/sherpa-onnx | Build the NVIDIA CUDA execution provider (requires the CUDA Toolkit; the nvcc host compiler is resolved from the installed toolkit, not hardcoded) | Build the NVIDIA CUDA backend (requires CUDA Toolkit and a compiler version supported by nvcc) |
| sci-ml/stable-diffusion-cpp | - | - |
| sci-ml/whisper-cpp | - | - |
| dev-libs/caffe | - | - |
| dev-python/nvidia-apex | - | Add support for CUDA processing |
| media-gfx/ideepcolor | - | - |
| net-misc/gminer-bin | - | Support Nvidia graphics cards |
| net-misc/lolminer-bin | - | Support Nvidia graphics cards |
| net-misc/srbminer-multi-bin | - | Support Nvidia graphics cards |
| net-misc/xmrig | - | Build cuda plugin |
| net-misc/xmrigcc | - | Build cuda plugin |
| app-accessibility/whisper-cpp | - | Enable NVIDIA CUDA GPU backend |
| media-gfx/stable-diffusion-cpp | - | - |
| media-video/ffmpeg | - | Enable support for various GPU-accelerated filters using NVIDIA PTX compiled with <pkg>llvm-core/clang</pkg> |
| app-misc/ollama | - | Enable NVIDIA CUDA support |
| dev-cpp/kokkos | - | Enable the CUDA execution space. Any Kokkos application compiled for CUDA embeds CUDA code via template metaprogramming. Thus, the whole application must be built with a CUDA-capable compiler. (At the moment, the only such compilers are NVIDIA’s NVCC and Clang 10.0+) |
| acct-user/ollama | - | - |
| app-containers/docker | - | - |
| dev-cpp/eigen | - | - |
| dev-libs/pocl | - | Enable the CUDA backend for NVIDIA GPUs |
| dev-libs/starpu | - | - |
| dev-util/bear | - | - |
| llvm-runtimes/openmp | - | - |
| media-libs/oidn | - | - |
| media-libs/openimageio | - | - |
| media-libs/opensubdiv | Enable NVIDIA CUDA Toolkit support through dev-util/nvidia-cuda-toolkit | Enable NVIDIA CUDA Toolkit support through <pkg>dev-util/nvidia-cuda-toolkit</pkg> |
| net-analyzer/suricata | - | Enable NVIDIA Cuda computations support |
| net-misc/sunshine | Enable accelerated video encoding on NVIDIA hardware | Enable accelerated video encoding on NVIDIA hardware |
| sci-chemistry/gromacs | Enable cuda non-bonded kernels | Enable cuda non-bonded kernels |
| sci-chemistry/vmd | - | - |
| sci-libs/cholmod | - | Use nvidia cuda toolkit for speeding up computations |
| sci-libs/clblast | Build with support for cuda instead of opencl (beta!) | Build with support for cuda instead of opencl (beta!) |
| sci-libs/dlib | Enable support for CUDA for Deep Neural Networks (cuDNN) on GPU (experimental) | Enable support for CUDA for Deep Neural Networks (cuDNN) on GPU (experimental) |
| sci-libs/flann | - | - |
| sci-libs/ginkgo | Add support for cuda assimp (dev-util/nvidia-cuda-toolkit) | Add support for cuda assimp (<pkg>dev-util/nvidia-cuda-toolkit</pkg>) |
| sci-libs/pastix | - | - |
| sci-libs/pcl | - | - |
| sci-libs/spqr | - | - |
| sci-libs/suitesparse | - | - |
| sci-libs/suitesparseconfig | - | - |
| sci-libs/trilinos | - | - |
| sci-libs/vtk | - | - |
| sci-mathematics/mathematica | - | Install with cuda support |
| sci-mathematics/petsc | - | - |
| sci-misc/boinc | - | - |
| sci-ml/caffe2 | - | - |
| sci-ml/ggml | - | - |
| sci-ml/gloo | - | - |
| sci-ml/kineto | - | - |
| sci-ml/pytorch | - | - |
| sci-ml/tensorpipe | - | - |
| sci-ml/torchvision | - | - |
| sci-physics/espresso | - | - |
| sci-physics/root | - | - |
| sys-apps/hwloc | Enable CUDA device discovery using libcudart | Enable CUDA device discovery using libcudart |
| sys-block/libfabric | - | - |
| sys-cluster/openmpi | - | - |
| x11-wm/xpra | - | - |
| sci-misc/ik_llama-cpp | - | - |
| sci-misc/stable-diffusion-cpp | - | - |
| sci-ml/ollama | Enable NVIDIA CUDA support | Build the CUDA (cuda_v13) llama-server GPU backend via <pkg>dev-util/nvidia-cuda-toolkit</pkg> |
| sci-libs/einspline | - | Build with cuda |
| app-misc/ollama-bin | - | - |
| media-gfx/luxcorerender | - | - |
| app-local-ai/audio-cpp | - | - |
| sci-ml/torchaudio | - | GPU acceleration via CUDA |
| sci-ml/torchcodec | - | Build the NVIDIA CUDA decoding/encoding backend (libnpp + libnvrtc) |
| sci-libs/libxc | - | - |
| sci-ml/comfy-kitchen | - | - |
| dev-python/NewsFrames | - | - |
| dev-python/NewsSentiment | - | - |
| dev-python/absl-extra | - | - |
| dev-python/ctransformers | - | - |
| dev-python/datadm | - | - |
| dev-python/deface | - | - |
| dev-python/elephant | - | - |
| dev-python/gpjax-nightly | - | - |
| dev-python/jax | - | - |
| dev-python/jaxutils | - | - |
| dev-python/libceed | - | - |
| dev-python/onnxruntime-gpu | - | - |
| dev-python/psyneulink | - | - |
| dev-python/py4DSTEM | - | - |
| dev-python/spacy | - | - |
| dev-python/spacy-nightly | - | - |
| dev-python/spacy-transformers | - | - |
| dev-python/spacy-wrap | - | - |
| dev-python/thinc | - | - |
| dev-python/cupy | - | Enable support for nVidia CUDA |
| sci-libs/lightgbm | - | - |
| dev-libs/boost-compute | - | build cuda examples and tests |
| sci-biology/fsl | - | Add support for CUDA |
| sci-biology/megahit | - | Enable NVIDIA GTX680 (4G memory) and Tesla K40c (12G memory) with CUDA 5.5, 6.0 and 6.5 |
| sci-chemistry/relion | - | Enable CUDA support |
| sci-libs/arrayfire | - | Build CUDA backend. |
| sci-libs/torchaudio | - | - |
| sys-devel/DPC++ | - | use the cuda backend |
| dev-cpp/thrust | - | - |
| dev-python/comfy-aimdo-bin | - | - |
| dev-python/comfy-kitchen-bin | - | - |
| dev-python/hyperspy | - | - |
| dev-python/sasmodels | - | Enable CUDA GPU-accelerated model evaluation via <pkg>dev-python/pycuda</pkg> |
| dev-python/tilelang | - | - |
| dev-python/vllm | - | - |
| dev-python/xgrammar | - | - |
| sci-libs/onnxruntime | - | Build the NVIDIA CUDA execution provider |
| sci-ml/bitsandbytes | - | - |
| sci-ml/flash-attn | - | - |
| sci-physics/lammps | - | Enable cuda gpu computing support |
| dev-python/comfy-aimdo | - | - |
| virtual/caffe2 | - | - |
| virtual/comfyui | - | - |
| virtual/pytorch | - | - |