{"library":"nvidia-nccl-cu11","type":"library","category":null,"description":"The `nvidia-nccl-cu11` package provides the NVIDIA Collective Communication Library (NCCL) runtime binaries specifically compiled for CUDA 11. NCCL is a high-performance library for collective communication operations (e.g., all-reduce, all-gather, broadcast) across multiple GPUs, both within a single node and across multiple nodes. It is optimized for NVIDIA GPUs and high-speed interconnects like NVLink and InfiniBand. This package primarily serves as a backend dependency for deep learning frameworks (like PyTorch, TensorFlow) and other GPU-accelerated libraries that require NCCL's capabilities for distributed computing. The current version is 2.21.5, with frequent updates corresponding to new NCCL releases and CUDA versions.","language":"python","status":"active","version":"2.21.5","tags":["GPU","deep learning","distributed computing","NVIDIA","CUDA","collective communication"],"install":[{"cmd":"pip install nvidia-nccl-cu11","imports":["from nvidia import nccl"]}],"homepage":"https://developer.nvidia.com/cuda-zone","github":null,"docs":null,"changelog":null,"pypi":"https://pypi.org/project/nvidia-nccl-cu11/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"compatibility":{"summary":{"python_range":"3.10–3.9","success_rate":50,"avg_install_s":4.5,"avg_import_s":null,"wheel_type":"wheel"},"url":"https://checklist.day/v1/registry/nvidia-nccl-cu11/compatibility"},"provenance":{"verified_status":"passing","verified_at":"Tue Jun 30","last_verified":"Tue Jun 30","next_check":"Thu Jul 30","install_tag":null}}