Tag: #RadixAttention
Kubernetes, DevOps and MLOps, written from things we ran and measured ourselves · 1 posts
Comparing vLLM and SGLang, and How to Benchmark Properly
We compare the two inference engines across prefix cache architectures, design philosophies, features, and selection criteria by workload shape, and review official tools for benchmarking methods including open-loop load
2026-10-07 · 40 min read #vLLM#SGLang#RadixAttention#벤치마크#용량 계획