Glimpse is Elorian's first AI model, a "visual thinking" model released on 16 October 2026 that reasons about pictures and video by going back to look at them while it thinks.
In plain words
Most models that read pictures turn an image into words and then reason in text, a chain Elorian calls fragile. Glimpse keeps returning to the picture: it can zoom into one area, watch a short stretch of video more closely, or follow one feature over time. It can also use tools that write code or edit images, and hand parts of a job to helper agents. It is a research preview, in an API and chat app, at $0.60 per million input tokens and $3.00 per million output tokens.
A quick example
Imagine you film a ball rolling off a table and ask a model when it starts to fall. A visual thinking model could go back to the frames near the table edge and study them closely before it answers.
Where you'll meet it
- Reading charts and lab data in science.
- Tagging short videos and spotting what a robot finished doing in its footage.
Related
Sources
- DocsIntroducing Glimpse, Elorian · read 16 Oct 2026
- DocsModels, Elorian · read 16 Oct 2026
- DocsElorian, Elorian · read 16 Oct 2026
Checked by WhatIsAI against the sources above · 16 Oct 2026. A full five-minute explainer with a diagram may follow. Spotted an error? Report it.