Elorian’s Glimpse model uses images to reason visually

Elorian, a Palo Alto AI startup, has released its first model, Glimpse, as a research preview with API access. Its chat interface is available by waitlist, according to TechCrunch, which reported the launch exclusively. The company raised a $55 million seed round at a $300 million valuation before launching a product.

Elorian says Glimpse can revisit images while reasoning, including by cropping into regions, extracting video frames, or running code to measure or compare visual details. The company calls this a “visual scratch space.” The approach is designed to let the model check what it sees rather than rely only on an initial image conversion to tokens.

Elorian says a small team trained Glimpse using a few hundred GPUs, and that the model is substantially smaller than the frontier systems in its evaluations. The company’s tests report that Glimpse outperforms GPT-6 Astra on scientific chart reasoning and short-form video captioning, with competitive results on other visual reasoning tasks. These are Elorian’s own evaluations.

Pricing is $0.60 per million input tokens and $3 per million output tokens. Elorian’s initial target customers are companies with large amounts of video, including robotics firms, scientific labs, and media and creative companies. CEO Andrew Dai said the company aims to build “the world’s best visual reasoning model” next year. Over the longer term, Elorian wants to use scene understanding to select actions, including controlling robots.