Optimized primitives for collective multi-GPU communication
- communications
- cpp
- cuda
- deep-learning
Repository Index
83,586 visible repositories in the index
1–4 of 4 · Most stars
Document scanning app
UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)
- ai
- allreduce
- amd
- broadcom
A few classes for extremely fast json parsing/serializing in modern C++. Possibly the fastest json parser in C++. Possibly the fastest json serializer in C++.
- cpp
- json
- json-parsing
- json-parsing-library