Excited to announce our work "Attention Drift" has been accepted to NeurIPS 2026 🎉
I'm so happy to see our method's widespread adoptation. Now supported by all major inference engines, vLLM, Tokenspeed, SGLang and TRT-LLM. Also community trained speculative decoding models for Minimax M3 (
@inferact and
@AMD) and Kimi K2.6-K3 trained (
@lightseekorg) being served in production as EAGLE 3.1
Shout out to all the great team mebers
@themishtimoi @yigitcancomlek @kaan_oktay @hongyangzh and Stephen Xia.
And special thanks to
@fal for helping for making this research possible 🙂