Standard Vision-language models (VLMs) reason about images and videos through language, powering a wide variety of applications from image captioning to visual question answering.

Autoregressive VLMs generate tokens sequentially, which prevents parallelization and limits… pic.twitter.com/54ahfojDZu

— Runway (@runwayml) 24 septembre 2025

Standard Vision-language models (VLMs) reason about images and videos through language, powering a wide variety of applications from image captioning to visual question answering. Autoregressive VLMs generate tokens sequentially, which prevents parallelization and limits

→ View original post on X — @runwayml