Databricks Closes $5B Round at $190B Valuation as Agent Demand Reshapes Enterprise AI

One of the largest private financings in software history just closed — and it happened because of AI agents. On August 13, 2026, Databricks announced it had closed a $5 billion strategic funding round at a $190 billion valuation, capping a month-long buying frenzy that began with a term sheet at a $188 billion valuation on July 16, saw investors arrive with roughly $15 billion of demand, and ended with the company taking a fraction of what the market offered — it had originally set out to raise only about $1 billion.

What Happened

The San Francisco-based “Data + AI” company’s round was led by existing investor Coatue, with participation from Blackstone, MGX, T. Rowe Price advisory accounts, and new investor Sixth Street Growth. The newcomer list also included BOND, Clearlake Capital, Point72, Premji Invest, and TPG, while the returning-shareholder roster reads like a who’s-who of growth investing: Andreessen Horowitz, Dragoneer, Fidelity, Franklin Templeton, GIC, Goldman Sachs Growth, Insight Partners, J.P. Morgan Private Capital, Morgan Stanley Investment Management, NEA, the Ontario Teachers’ Pension Plan, Temasek, Thrive Capital, and WCM.

The final $190 billion valuation represents a roughly 40% jump from the $134–135 billion range in late 2025, achieved barely two months after the company first signaled the raise. And per TechCrunch, the supply-demand dynamics were extraordinary: Databricks wanted to raise about $1 billion, investors wanted in for close to $15 billion, and the company settled on $5 billion — deliberately leaving money on the table to control dilution while still arming itself for product investment.

The Numbers Behind the Valuation

This is not a speculative mark. Databricks’ announced Q2 figures include:

• Revenue run-rate surpassing $7 billion, growing more than 80% year-over-year
• Positive adjusted free cash flow — the company is not burning cash even at hypergrowth
• Lakehouse data warehousing surpassing a $1.5 billion revenue run-rate, growing over 100% YoY
• Lakebase, its serverless Postgres database built for AI agents, crossing a $100 million revenue run-rate
• More than 1,000 customers spending $1M+ annually, over 100 of which spend more than $10M

Over 20,000 organizations worldwide use the platform, including adidas, AT&T, Bayer, Block, Mastercard, Rivian, Unilever, and 70% of the Fortune 500. Growing 80% at a $7 billion revenue scale is nearly unprecedented in enterprise software history — most companies at that size have decelerated into the 20–40% range.

Why AI Agents Rewrote the Script

The strategic logic of the round revolves around a new kind of customer: not human users, but AI agents. In its press release, Databricks positions its platform as the substrate agents run on — reliable, scalable, real-time data infrastructure, plus the governance to control which models agents use and how much budget they burn.

Three products anchor that thesis:

1. Lakebase — a serverless Postgres database built for AI agents, which crossed a $100M run-rate barely a year after launch. Agents need low-latency transactional state, and Postgres compatibility slashes switching costs.
2. Genie — Databricks’ “AI coworker” that turns business data into trusted answers and actions, supplying the enterprise context raw models lack.
3. Unity AI Gateway — a control plane for multi-model governance, intelligent routing, and cost management, aimed squarely at the most painful enterprise problem of the moment: agents burning expensive tokens at scale.

“Enterprises don’t want AI that just talks. They want agents that operate in their business, remember context, give accurate answers, and don’t blow the budget,” said Ali Ghodsi, co-founder and CEO of Databricks. “That requires Lakebase for real-time operational data, Genie for context across the enterprise, and Unity AI Gateway for multi-model cost control. The overwhelming response to this round is proof our AI strategy is winning.”

Coatue co-founder Thomas Laffont added: “Databricks has been early to where AI is heading for a decade. Today it is infrastructure for the entire industry to build and scale AI. The most impressive thing is the speed — compressing multi-year R&D timelines into months, operating more like a research lab than a typical software company.”

The AGI Footnote

The financing news came with a provocative footnote: per Forbes, Ghodsi has been telling investors that AGI has “already arrived” — not as a single superintelligence, but as the accumulated ability of agentic systems wired into real data to do real work. Whether or not one accepts the framing, it reveals Databricks’ positioning strategy: as that Forbes headline put it, the fortune is now buried in “context” — and Databricks intends to own the context layer.

What It Means

The round is the clearest signal yet that AI investment is rotating from the model builders to the infrastructure suppliers. While frontier labs raise ever-larger sums to train models, Databricks monetizes the layer every deployment ultimately needs: data pipelines, governance, and cost control. Its ability to stay free-cash-flow positive at 80%+ growth also strengthens the argument that there is real, durable enterprise revenue under this AI cycle — in sharp contrast to the zero-revenue startups Ghodsi has publicly criticized.

Competitive pressure is real — Snowflake, Microsoft (which expanded its Databricks partnership in July to bring commercial context to enterprise AI), and every hyperscaler want the same agent workloads. But with a $190 billion valuation, $5 billion in fresh capital, and a product portfolio purpose-built for the agent era, Databricks now sits among the most valuable private companies in the world — and one of the few whose growth curve is still nearly vertical.