cd/entity/CUDA graphs· home› entities› CUDA graphs
grep -l @cuda graphs /news/*.json | wc -l → 1

CUDA graphs

mentions 1 type Person feed RSS

// recent coverage 1 mentions

18:06
2026-10-07
zhebrak.io
ai-infrastructure

Decode Decoded

A benchmark of vLLM 0.30.0, SGLang 0.5.20, and TensorRT-LLM 1.2.1 on a single H100 SXM GPU with 8 CPU cores found no host-bound regime across Qwen 3 0.6B-32B models at batch sizes from 1 to 256, with …

// co-occurs with top 7 entities
vLLM 1 SGLang 1 TensorRT-LLM 1 NVIDIA H100 SXM 1 Qwen 3 1 Nsight Systems 1 Model Runner V2 1
// topics top 3 topics
ai infrastructure 1 large language models 1 mlops 1