Santiago says a model called Glimpse can answer questions about an image or video, including how objects relate spatially and how an action changes over time. Andrew Dai introduces Glimpse as a visual thinking model with competitive frontier performance against Astra on key public visual reasoning benchmarks.
You can show this model an image or a video, and it will answer any questions about it.
It goes way beyond what frontier models can do.
For example, Glimpse can tell you how objects in an image relate to each other, or how an action changes over time in a video.
Robotic teams would kill for these capabilities!
Spent years watching models ace a single image and completely whiff on anything that moves or changes angle. Today we’re fixing that. Try Glimpse now, our first visual thinking model with competitive frontier performance compared to Astra on key public visual reasoning benchmarks.
Source: Santiago · x.comPublished · added here
