Elorian AI, a startup founded by former Google DeepMind and Apple researchers, secured a $55 million seed round at a $300 million valuation before shipping any product, betting that AI should reason directly in visual space rather than translating images into text first.
April 9, 2026 · Elorian AI
$55M Raised, Zero Products Shipped
Ex-Google DeepMind and Apple researchers land a seed round at a $300M valuation — before any model, API, or revenue — on a single thesis: visual reasoning precedes language.
$300M
Valuation, pre-product
~5 mo
From leaving Google to close
$80T
Physical economy targeted
Raised on pedigree — funding vs. output
A near-vertical bet: the money arrived on team and thesis alone — no public model, API, or revenue.
The thesis · visual-first vs. text-first
Today's vision-language models
Image → Text → Reason
Weak on spatial relationships, physical constraints and design intent.
Elorian's approach
Image → Reason directly in visual space
Models operate on representations that encode structure, relationships and constraints.
The bull case
Team drawn from Google Brain & DeepMind; CEO led Gemini data efforts
Native visual understanding could unlock robotics, engineering, architecture
Backers cite work on a "verifier" for visual reasoning
Claims early wins over Gemini 3 Pro on select benchmarks
The skeptics
No public model, API or product to test
Benchmark claims remain unverified
3D structure & physical grounding unproven until launch
Competing against frontier systems like Gemini and GPT
A state-of-the-art model is promised within 2026.
Whether it delivers on the benchmark claims will decide if the $300M valuation looks prescient or premature .
Continue reading The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in ✓ Signed in — this article isn’t included in your current plan.Unlocking the full article…