NCCL 2.31.2 ships GPU-driven CFT/RMA, 0-SM collectives, better multi-NIC for Blackwell-scale training
SkyLi0n · x · 2026-09-23
NCCL 2.31.2 is out with GPU-driven CFT/RMA for fine-grained communication, per-collective tuning, NVLS+PAT improvements, 0-SM collectives, improved multi-NIC/GIN support, and PACE fusion. The author sees it as useful for MoE, FSDP, TP/EP, and Blackwell-scale training.
Related event: NVIDIA Ships NCCL 2.31.2 with GPU-Driven Communication and 0-SM Collectives(2 posts)→
More from Infra
- Gas-for-AI-datacenters forum co-located with CoRL shows energy demand spillover — carlosdponx · 2026-11-12
- a16z podcast: Mike Volpi on Cisco's 70+ acquisition playbook and the AI buildout — a16z · 2026-11-12
- AMD to pass Intel in server revenue by mid-2027 as share gains 6 points a year — BenBajarin · 2026-11-12
- Study: Reflective Data Center Roofs Actually Cool Daytime Ground by 1.8-2.2°C — SumitGup · 2026-11-12
- Tsinghua's BaGuaLu-2 tames MoE expert-load spikes at runtime, accepted to SC26 — jiqizhixin · 2026-11-12
- antirez backs low-bit DeepSeek 4.1 Flash over Qwen 27B dense on Apple M5 Macs — antirez · 2026-11-12