{"id":23477,"library":"cuequivariance-ops-torch-cu13","title":"cuequivariance-ops-torch (CUDA 13)","description":"GPU-accelerated PyTorch extensions for equivariant neural network primitives, specifically compiled for CUDA 13. Current version 0.10.0, requires Python >=3.10. Part of the cuequivariance ecosystem for efficient group-equivariant operations.","status":"active","version":"0.10.0","language":"python","source_language":"en","source_url":"https://github.com/QUVA-Lab/cuequivariance","tags":["equivariant","torch","CUDA","geometric deep learning","group theory"],"install":[{"cmd":"pip install cuequivariance-ops-torch-cu13","lang":"bash","label":"Install for CUDA 13"}],"dependencies":[{"reason":"PyTorch with CUDA support is required at runtime.","package":"torch","optional":false},{"reason":"Core library for equivariant operations.","package":"cuequivariance","optional":false}],"imports":[{"note":"The package name for import does not include the CUDA suffix; import using 'cuequivariance_ops_torch'.","wrong":"from cuequivariance_ops_torch_cu13 import ...","symbol":"all relevant ops","correct":"import cuequivariance_ops_torch as coe_ops"}],"quickstart":{"code":"import torch\nimport cuequivariance_ops_torch as coe_ops\n\ndevice = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\nx = torch.randn(10, 10, device=device)\ny = coe_ops.some_op(x)  # Replace with actual operation\nprint(y.shape)","lang":"python","description":"Basic usage: import the package (without CUDA suffix) and call an operation on a CUDA tensor."},"warnings":[{"fix":"Use 'import cuequivariance_ops_torch as coe_ops' in code.","message":"Import the package as 'cuequivariance_ops_torch', not with the CUDA suffix ('cuequivariance_ops_torch_cu13'). The suffix is only for pip installation to select the correct CUDA version.","severity":"gotcha","affected_versions":"all"},{"fix":"Match the package suffix (e.g., cu12, cu13) to your installed CUDA toolkit version. Use 'pip install cuequivariance-ops-torch-cu12' for CUDA 12.","message":"CUDA compatibility: The cu13 wheel only works with CUDA 13.x. Installing on a system with a different CUDA version will cause runtime errors (e.g., 'CUDA driver version is insufficient').","severity":"breaking","affected_versions":"0.x"},{"fix":"Ensure your GPU is supported (e.g., Ampere or newer). If not, you may need to build from source or use a CPU fallback.","message":"The package may not be compatible with all CUDA compute capabilities; check the wheel's target architectures (e.g., sm_80, sm_90). Running on an unsupported GPU will raise a 'no kernel image' error.","severity":"gotcha","affected_versions":"all"}],"env_vars":null,"last_verified":"2026-05-01T00:00:00.000Z","next_check":"2026-07-30T00:00:00.000Z","problems":[{"fix":"Change import to 'import cuequivariance_ops_torch' (without the CUDA suffix).","cause":"Trying to import directly from the pip package name instead of the importable module name.","error":"ImportError: cannot import name '...' from 'cuequivariance_ops_torch_cu13'"},{"fix":"Use a version compiled for your GPU's compute capability, or build from source. Alternatively, set environment variable 'CUDA_LAUNCH_BLOCKING=1' for debugging.","cause":"The installed binary was compiled for a different GPU architecture (e.g., sm_80) and is not compatible with your GPU.","error":"RuntimeError: CUDA error: no kernel image is available for execution on the device"},{"fix":"Run 'pip install cuequivariance' to install the base dependency.","cause":"The core 'cuequivariance' library is not installed.","error":"ModuleNotFoundError: No module named 'cuequivariance'"},{"fix":"Install a CUDA-enabled version of PyTorch, e.g., 'pip install torch==2.x+cu118' (adjust to your CUDA version).","cause":"PyTorch is installed without CUDA support (CPU-only), but the operations require GPU.","error":"AssertionError: Torch not compiled with CUDA enabled"}],"ecosystem":"pypi","meta_description":null,"install_score":null,"install_tag":null,"quickstart_score":null,"quickstart_tag":null}