Elorian introduced Glimpse on October 16 as a research-preview model for understanding images, video and physical events. It can inspect visual evidence repeatedly, follow objects and reason about spatial relationships.
The company offers API and chat access at $0.60 per million input tokens and $3 per million output tokens. It says the model was post-trained from an open-weight base, which does not establish that Glimpse’s own weights are publicly released.
Elorian’s comparisons use specified thinking effort and agent tools. They should be read as benchmark results under those settings, rather than a universal “best visual model” ranking.
Sources: Elorian announcement