Inco AI(@inco_ai):DFlash drafts a whole block in parallel. DFlash 2 picks one path through the block instead of the top token at each position: over 20% more tokens per pass, about 1% more latency. Our co-founder and chief scientist @zhijianliu explains how at @modal Runtime in SF on Oct 1.

Inco AI

@inco_ai

Inference, reimagined for the agentic era

加入 May 2026

21 正在关注 3.9K 粉丝

Inco AI@inco_ai

2026.09.24 21:00

DFlash drafts a whole block in parallel. DFlash 2 picks one path through the block instead of the top token at each position: over 20% more tokens per pass, about 1% more latency. Our co-founder and chief scientist @zhijianliu explains how at @modal Runtime in SF on Oct 1.

显示更多