SGLang Optimizes GLM 4.5 Inference to 535 token/s
BanghuaZ · x · 2026-07-15
The SGLang team continues to optimize the inference performance of the GLM 4.5 model, with the latest version achieving a throughput of 535 token/s. This progress was made possible through close collaboration with NVIDIA and Zhipu AI (Z.ai).
More from Infra
- AWS on agent prompt caching: explicit + implicit patterns to slash token costs — AWS ML Blog · 2026-10-27
- Thieves spoofing GPS steal $100M+ in data center cargo since June — Ars Technica AI · 2026-10-27
- Lam Research: weekly token usage up ~250x in 18 months — Beth_Kindig · 2026-10-27
- AWS shows how to build enterprise agentic retrieval on Bedrock with CDK — AWS ML Blog · 2026-10-27
- Encrypted GPT-2 speedrun challenges FHE inference: best is 3.031 s/token, goal sub-1s — tensorqt · 2026-10-27
- Open-source Latch compiles PyTorch into training megakernels, 2.75x faster on H100 — sguada · 2026-10-27