Devin Maker Cognition Closes $2B Round at $48B, Nearly Doubling in Four Months

Cognition, the company behind AI programmer Devin, closed a $2 billion funding round on September 8, 2026 at a $48 billion valuation — up from the $26 billion valuation it set just four months earlier.

From Rumored $40 Billion to a Confirmed $48 Billion

This closed round confirms and exceeds the roughly $40 billion figure that had reportedly been under discussion in prior weeks — a specific signal that investor demand for the round exceeded even the elevated valuation initially being floated, rather than the deal closing at or below early expectations.

Investor Syndicate

The round was led by Andreessen Horowitz, Accel, Founders Fund, General Catalyst, and Avenir — a roster of some of the most prominent venture firms in enterprise and AI investing.

Rapid Revenue Growth

Since its last round in May 2026, Cognition's annualized revenue run rate has grown from $492 million to $900 million — nearly doubling in roughly four months — with the company reportedly on track for a $4-5 billion annualized run rate by the end of 2026, according to The Information.

Enterprise Clients

Major enterprise clients include Mercedes-Benz, NASA, and Goldman Sachs — a customer list spanning automotive, government/aerospace, and financial services, indicating broad enterprise appeal well beyond software-native industries.

One of Several Massive AI Coding Valuations This Month

This valuation lands just one week before rival Factory tripled its own valuation to $5 billion, underscoring how much capital is concentrating specifically into the AI coding agent category this quarter, rather than being spread evenly across the broader enterprise AI landscape.

What This Means for the AI Coding Agent Market

Investors backing multiple, well-capitalized AI coding agent companies at rapidly escalating valuations within the same month is a specific signal that this market is viewed as large enough to support multiple major winners, rather than a single winner-take-all outcome — a useful framing for enterprises currently evaluating which AI coding agent vendor to standardize on, since consolidation risk in this category may be lower than in more typically winner-take-all software markets.