The benchmarks show the gap. NVLS all-reduce latency drops from 586.1µs on H200 to 313.3µs on GB200....
Perplexity(@perplexity_ai)107 字 (约 1 分钟)
85
NVLS all-reduce latency significantly improves from 586.1µs on H200 to 313.3µs on GB200, with notable performance gains in MoE prefill and decode throughput.
入选理由:NVLS all-reduce latency drops from 586.1µs on H200 to 313.3µs on GB200.
FeaturedTweet#NVLS#H200#GB200#MoE#Performance英文
