# Decagon shares lessons on speculative decoding with DFlash and EAGLE

> Original title: Nicholas Liu (@DecagonAI) on speculative decoding in practice.

- Source: LMSYS Org
- Published: 2026-11-09T23:15:44.000Z
- Bad Influence: https://badinfluence.news/items/ifk6m05ca2il6487idiuvyb0e
- Original: https://x.com/lmsysorg/status/2119931978367774816

## Summary

Nicholas Liu of Decagon shares lessons from evaluating DFlash and EAGLE for speculative decoding. The post says concurrency, MoE verification costs and structured outputs shape latency, cost and deployment decisions.

## Post

Nicholas Liu (@DecagonAI) on speculative decoding in practice.

Lessons from evaluating DFlash and EAGLE: how concurrency, MoE verification costs and structured outputs shape latency, cost and deployment decisions.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWuCmXhaMAASew_.jpg&mode=thumb&exp=1794528000&sig=85f82f80c848d057)