NVIDIA

NVIDIA/nccl

Optimized primitives for collective multi-GPU communication

1 good first / help-wanted issue · C++ · last activity Oct 28, 2026

5.2K stars 1.5K forks 5.2K watchers C++ Other
communications cpp cuda deep-learning gpu nvidia
View on GitHub Website
1 Open Issue Need Help Last updated: Oct 28, 2026

Open Issues Need Help

View All on GitHub
[RFC][Help Wanted] Reduce NCCL's GPU memory footprint about 2 hours ago
enhancement help wanted Community Discussion
NVIDIA/nccl
5.2K

Optimized primitives for collective multi-GPU communication

C++
#communications#cpp#cuda#deep-learning#gpu#nvidia