Elorian, a Palo Alto AI startup that raised a $55 million seed round at a $300 million valuation before shipping a product, released its first model, Glimpse, on Friday as a research preview. The model has API access, while chat access is waitlist-only, so Elorian is currently offering results tables and a thesis rather than a finished product.

The thesis is that today’s multimodal models don’t truly see: they look at an image once, convert it to tokens, and reason about it in text, like describing a photo from memory. Glimpse instead returns to the visual source during reasoning. It can crop into a region, pull specific frames from a video, or run code to measure or compare what it’s looking at, a mechanism Elorian calls a “visual scratch space.” CEO Andrew Dai, formerly of Google DeepMind, described photographing a finished board game and asking a frontier model who was winning; the model couldn’t answer, even though the board’s symbols were clear to an adult. He drew the same contrast with Ikea diagrams, which humans can follow with no text at all.

The approach is not entirely new — OpenAI added crop-and-zoom during reasoning to its o-series models in 2025 — but Elorian says it built the entire model around this behavior, and did so cheaply. A small team trained Glimpse on a few hundred GPUs, and the model is substantially smaller than the frontier systems it is benchmarked against. Glimpse is priced at $0.60 per million input tokens and $3 per million output tokens, cheaper than mid-level models like Sonnet and Sol but more expensive than lightweight models like Haiku and Luna.

Elorian‘s own evaluations claim Glimpse outperforms GPT-6 Astra on scientific chart reasoning and short-form video captioning, with competitive results on other visual reasoning tasks. The company came out of stealth six months ago with a round co-led by Striker Venture Partners, Menlo Ventures, and Altimeter Capital, with participation from Nvidia and Google chief scientist Jeff Dean. Initial targets are companies with large amounts of video: robotics firms, scientific labs, media, and creative companies. Longer term, Elorian wants to move from seeing to doing, building a model of a scene and using it to choose an action, including controlling robots. Dai said the company aims next year to build “the world’s best visual reasoning model,” one that leads on every visual reasoning test.