Elorian的首个AI模型旨在与图像进行推理，而不仅仅是关于图像推理。

Elorian’s first AI model wants to reason with images, not just about them | TechCrunch

Last day to save up to $200 to TechCrunch Founder Summit on Oct. 16. Register now.

Disrupt doors open Oct. 13. Get your pass and bring someone with you at 50% off. REGISTER NOW.

Image Credits:Build Mode podcast

Elorian’s first AI model wants to reason with images, not just about them

Elorian, the Palo Alto AI startup that raised a $55 million seed round at a $300 million valuation before it launched a product, released its first model on Friday, TechCrunch has exclusively learned.

The model, Glimpse, is launching as a research preview with API access and a chat interface is waitlist-only, so for now its a results table and a thesis. The thesis? Today’s AI models, for all their skill with language and code, still can’t really see.

Most multimodal models look at an image once, convert it to tokens, and then reason about it in text — like describing a photo from memory. Glimpse keeps going back to the source. Mid-reasoning, it can crop into a region, pull specific frames from a video, or run code to measure or compare what its looking at, a mechanism Elorian calls a visual scratch space. Think of someone pinching to zoom on a phone photo to double-check what they thought they saw.

For former Google DeepMind researcher and Elorian CEO Andrew Dai, one of the clearest signs that AI still couldn’t really see came on game night.

Last year he photographed a board game hed just finished with friends and asked a frontier model who was winning. It couldnt say. Board games each have their own visual language, each have their own iconography, symbols, he told TechCrunch. To an adult, its quite easy to understand whats going on with minimal text, but these models arent able to intuit what the symbols mean.

Humans can look at Ikea diagrams, with no text, and assemble Ikea furniture, right? Why cant AI models? he said.

The approach isnt completely new — OpenAI added crop-and-zoom during reasoning to its o-series models in 2025 — but Elorians pitch is that it built the whole model around it, cheaply. The company says a small team trained Glimpse on a few hundred GPUs and that the model is substantially smaller than the frontier systems its benchmarked against.

Glimpse costs $0.60 per million tokens in and $3 per million tokens out — cheaper than mid-level models like Sonnet and Sol but more expensive than lightweight models like Haiku and Luna.

But according to Elorians evaluations, Glimpse outperforms even GPT-6 Astra on scientific chart reasoning and short-form video captioning, with competitive results on other visual reasoning tasks.

Elorian 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.

The initial targets are companies sitting on lots of video: robotics firms, scientific labs, media, and creative companies. Longer term, Elorian wants to move from seeing to doing. That means building a model of a scene using that to choose an action, including directly controlling robots. Next year, Dai said, the company aims to build “the world’s best visual reasoning model,” one that leads on every visual reasoning test.

AI, ai models, Exclusive, Startups, TC

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

Aditya Mehta is a reporter at TechCrunch covering AI. Hes supported by the Tarbell Center for AI Journalism and attended UC Berkeley. You can contact from Aditya by emailing [emailprotected] or via encrypted message at adymehta.74 on Signal.

Cover more ground at TechCrunchs ultimate founder bootcamp of the year by bringing your co-founder or partner. Save 50% off a second pass. Meet scaling and investment leaders, ask questions, find solutions, and more.

Tesla reveals SpaceX-powered Roadster that flies and falls with style

FBI arrests OpenAI employee accused of planting hidden recording device in conference room

Bumble agrees to a $1.3B acquisition deal as online dating faces a reset

Just 5 years in, interactive chatbot platform Character.AI files for bankruptcy

Meta releases its first frontier model, Muse Wonder

This AI startup says it can generate permit-ready blueprints in minutes

Elon Musk intensifies attack on Ambani over Starlink India launch delay