# SGLang reports 2.11–4.28× GLM-5.3-Flash throughput gains on H200 GPUs

> Original title: New blog: 2.11–4.28× Higher GLM-5.3-Flash Serving Throughput on H200

- Source: LMSYS Org
- Published: 2026-11-09T19:42:58.000Z
- Bad Influence: https://badinfluence.news/items/uajone8mg8p0gwde6bx0ltysi
- Original: https://x.com/lmsysorg/status/2119878431831978221

## Summary

Ant International AI Infra reports that its long-context serving optimizations for SGLang raise GLM-5.3-Flash throughput on H200 GPUs. On 4×H200, prefill throughput rises 2.06× for 64K-token inputs without a shared prefix, and decode throughput rises 4.32× for 64K/1K requests with 99.9% shared prefix. On 16×H200, end-to-end throughput rises 2.11–4.28× across three prefix-sharing ratios.

## Post

New blog: 2.11–4.28× Higher GLM-5.3-Flash Serving Throughput on H200

Ant International AI Infra brings faster long-context serving to SGLang. How much do cache reuse, speculative decoding, and kernel tuning add up to?
Engineering Result:
- 2.06× prefill throughput on 4×H200: 64K/10 tokens, no shared prefix.
- 4.32× decode throughput on 4×H200: 64K/1K tokens, 99.9% shared prefix.
- 2.11–4.28× end-to-end throughput on 16×H200: 64K/1K tokens across three prefix-sharing ratios.

Thanks to Ant International AI Infra and the SGLang community.

Read the full blog below

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWtSJCBaQAAZhE2.jpg&mode=thumb&exp=1794528000&sig=e87c2091de3268c3)