Recommended Articles
Fundraising

Applied Compute Raises $350 Million at $3.25 Billion Valuation for Enterprise AI

By Angeled · November 4, 2026 · Fundraising
Applied Compute Raises $350 Million at $3.25 Billion Valuation for Enterprise AI
Applied Compute Raises $350 Million at $3.25 Billion Valuation for Enterprise AI

Applied Compute, a San Francisco-based AI cloud and infrastructure startup, has raised $350 million in funding at a $3.25 billion valuation, the company announced on November 2, 2026, with coverage by SiliconANGLE and other outlets following on November 3 and 4. Prominent angel investor Elad Gil led the round, with participation from Nvidia, Kleiner Perkins, Lux Capital, Hanabi Capital, Blackstone, Shopify Ventures, HubSpot Ventures, and more than a dozen other investors.

In its funding announcement, the company disclosed that its annualized recurring revenue has topped $100 million — a milestone reached less than 18 months after the company was founded. The company counts Microsoft and DoorDash among its customers, according to SiliconANGLE's report. The new capital is earmarked for expanding its market presence and buying more infrastructure.

What Applied Compute Does

Applied Compute helps enterprises build, train, and own their custom AI models. Founded in 2025 by former OpenAI technical researchers Yash Patil, Rhythm Garg, and Linden Li, the company focuses on custom open-source AI models for enterprises, spanning training, inference, and ongoing model improvement.

That positioning sits between two familiar poles of the AI market. On one side are the frontier labs that build general-purpose models and sell access to them. On the other are the cloud providers that rent raw computing power. Applied Compute's argument is that many enterprises want something in between: models tailored to their own data, workflows, and industry, which they can control and keep improving, rather than renting a generic model or building infrastructure from scratch.

The distinction between training and inference matters here. Training is the expensive, one-time process of teaching a model; inference is the ongoing process of running it to generate outputs. Enterprises increasingly need both, plus the tooling to evaluate, update, and govern models once they are in production. Applied Compute sells that full lifecycle as a service.

A Valuation Ladder Built in 17 Months

The round is the third step in a remarkably steep valuation climb. The company first raised $20 million from Benchmark's Victor Lazarte at a $100 million valuation in mid-2025, then an $80 million round led by Kleiner Perkins at $1.3 billion. The latest round lifts that figure to $3.25 billion — roughly 32 times its annualized revenue run rate.

That multiple is a growth multiple, not a margin multiple, and it only makes sense if the revenue curve keeps bending upward. Supporters would argue that enterprise AI infrastructure is a winner-take-most market where early scale compounds: more customers mean more data, better tooling, and stronger references, which win more customers. Skeptics would note that an annualized run rate is a pace extrapolated across a year, not a completed year's revenue, and that the company has disclosed no profitability measure.

Why Nvidia's Participation Matters

Nvidia's participation is the detail rivals will watch most closely. Nvidia makes the chips that most AI infrastructure runs on, and its investment arm has become a strategic kingmaker in the AI ecosystem, backing companies that drive demand for its hardware and software platforms.

For Applied Compute, Nvidia's backing is both a vote of confidence and a practical advantage: closer alignment with the dominant chipmaker can mean earlier access to hardware, better optimization, and credibility with enterprise buyers. For Nvidia, investing in a company that helps enterprises train and serve open-weight models expands the ecosystem of workloads that run on its chips, rather than leaving that demand to a handful of closed-model providers.

Why It Matters for Founders and Investors

For founders, Applied Compute's trajectory carries a blunt lesson about speed. The company reached $100 million in annualized revenue in under 18 months — a pace that would have been extraordinary in any era and is nearly unheard of for enterprise infrastructure. That speed came from selling into a market where demand already existed and was urgent: enterprises that know they need custom AI but lack the teams and infrastructure to build it themselves.

For investors, the deal illustrates how the AI infrastructure market is stratifying. The largest rounds are concentrating in companies with demonstrated revenue, strong technical founding teams, and strategic investors. The $3.25 billion valuation also raises the bar for the next round: Applied Compute will need to convert its run rate into durable, retained enterprise revenue and show a path to margins that justify the multiple. If it does, it becomes a credible candidate for a future public listing; if growth stalls, the valuation will be difficult to defend.

Angeled.com connects founders with investors.

Sources: SiliconANGLE (November 3, 2026); FinSMEs (November 4, 2026).

Read Next