# Zhijian Liu explains DFlash and other speedups behind Inco's open-model endpoints

> Original title: Zhijian Liu (@inco_ai) on DFlash and the other optimizations behind Inco’s open-model endpoints.

- Source: LMSYS Org
- Published: 2026-11-09T23:15:43.000Z
- Bad Influence: https://badinfluence.news/items/x1za7nthdxr1agwbn0pnhx0rz
- Original: https://x.com/lmsysorg/status/2119931972499972463

## Summary

Zhijian Liu of Inco describes DFlash, a block-diffusion drafter now available in SGLang, and other optimizations behind Inco's open-model endpoints. The post also covers where the rest of the speed comes from.

## Post

Zhijian Liu (@inco_ai) on DFlash and the other optimizations behind Inco’s open-model endpoints.

A look at its block-diffusion drafter, now in SGLang, and where the rest of the speed comes from.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWuCY9HaUAAITtt.jpg&mode=thumb&exp=1794528000&sig=65677cd29d70162f)