Elorian releases Glimpse, a research preview model that reasons over images and video
Elorian introduced Glimpse on Oct 16, a research preview of what it calls a visual thinking model. Elorian says it follows objects in motion and across changing viewpoints, and connects what it sees to what happens next. Glimpse is available through Elorian's API and chat interface, at $0.60 per million input tokens and $3.00 per million output tokens. Elorian says a small team built it with a few hundred GPUs, post-trained on an open weights base model. Its benchmark comparisons with larger models are Elorian's own.
How you’d use it: Ask Glimpse through the API to count or track objects in a short video.