Elorian AI's Glimpse now available on AI Gateway
By Steven Van ·
Elorian's visual reasoning model takes images and video through AI Gateway, using the model name elorian/glimpse with the AI SDK or Chat Completions.
Elorian's Glimpse, a visual reasoning model, is now available on Vercel's AI Gateway, according to the release note. It accepts images and video and is meant to follow objects through motion, interpret spatial relationships, and connect an action to its physical consequences.
What Glimpse does
Glimpse can return to visual evidence while it works. It can zoom into a detail, revisit a short interval of video, or track the same feature across frames. Vercel lists scientific chart analysis, video captioning, visual counting, and robotics data annotation as use cases.
The examples given are checking a train's motion to determine whether it is slowing down, keeping objects distinct through camera movement, and labeling completed actions in robot footage.
Using it in code
The model name is elorian/glimpse. The announcement shows an AI SDK streamText call that sends an image as a file part alongside a text prompt, with maxOutputTokens set to 4096 and reasoning set to 'low'. A Chat Completions example is also included.
With AI SDK 7, a video can be sent as a file part too. The URL should be a public or signed HTTPS URL. The video example passes video_fps through providerOptions under elorian.
Setup and testing
To configure AI Gateway in Claude Code, Codex, Cursor, and other tools, install the latest Vercel CLI and run its setup. Glimpse can also be tried in the model playground.
AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations.