# MMFlash: Faster Speculative Decoding for Vision-Language Models

> Source: <https://aiflash.com/news/152287/>
> Published: 2026-11-16 06:30:00+00:00

Speculative decoding accelerates generation by letting a lightweight drafter propose tokens for parallel verification by a target model. Diffusion drafters such as DFlash propose an entire token block in a single forward pass, but mainly focus on text-only tasks, leaving multimodal drafting underexp
