From the source
Understanding begins when a model knows what deserves a second look.
Glimpse is a new category of visual thinking model: a system that reasons about the physical world from what it sees.
Glimpse follows objects in motion and across changing viewpoints, understands how parts fit together, and connects an observation to what happens next.
That understanding matters in robotics, scientific discovery, and the creative industries.
Compared to existing frontier large language models, Glimpse pushes furthest on bringing visual representations into the reasoning process itself.
It can reason in visual space, revisiting and refining internal visual representations as thinking unfolds, and is more grounded, learning to reason about objects and their spatial relationships.
As an agent, Glimpse is also trained end-to-end to use multimodal tools that write code, transform images, and watch videos; and Glimpse can spawn and coordinate subagents that exchange text, visual artifacts, and internal representations to solve a shared problem.
Together, these capabilities let Glimpse inspect a problem and revise its interpretation as it reasons.
…