Databricks' $190 Billion Round Shows Where Enterprise AI Money Is Flowing

Week of August 10, 2026

Databricks Run-Rate Passes $7 Billion as It Raises at $190 Billion

Databricks says its warehousing business and Lakebase database are both past key revenue milestones as AI agents drive platform demand.

View the data

	Run-rate threshold passed
Total revenue run-rate	$7B
Data warehousing	$1.5B
Lakebase	$0.1B

More charts in By the Numbers

Databricks raised $5 billion at a $190 billion valuation, the company said on August 13, and disclosed that its revenue run-rate has passed $7 billion with more than 80% year-over-year growth in its second quarter. Reuters reported the same figures that day. The round, which Databricks said was led by Coatue, was the largest of the week, according to Crunchbase News. For enterprise leaders, the more useful signal is what Databricks says customers are paying for: not models, but the data, context and cost controls that let AI agents do real work.

What did Databricks announce?

Beyond the headline numbers, Databricks said it had positive adjusted free cash flow over the last 12 months. It reported that Lakebase, its serverless Postgres database pitched as a foundation for AI agents, has passed a $100 million run-rate, and that its data warehousing business has passed $1.5 billion, growing more than 100% a year. More than 1,000 customers spend at a run-rate above $1 million, and more than 100 above $10 million.

Other participants include Blackstone, MGX, accounts advised by T. Rowe Price and new investor Sixth Street Growth. Databricks said it will use the money to build out Lakebase, Genie, which it describes as an AI coworker that turns business data into answers and actions, and Unity AI Gateway, which handles routing and cost controls across multiple AI models.

The jump is steep. In December 2025, Databricks said it was raising more than $4 billion at a $134 billion valuation, with a run-rate above $4.8 billion. Crunchbase News reported that the company has raised roughly $25 billion since its Series A in 2013.

Why are investors paying up for data platforms?

Databricks frames its strategy around a simple claim: companies now have a new kind of worker to support, AI agents, and those agents need reliable operational data, business context and predictable budgets. "Enterprises don't just want AI that talks," CEO Ali Ghodsi said in the release, arguing they want agents that remember context, answer accurately and finish work within budget.

That argument lines up with what practitioners have told AI Connect audiences for two years. Agents fail less often because the model is weak than because the data underneath is stale, poorly described or ungoverned. AI Connect panels on the emerging context layer and on lakehouse governance and cost reached the same conclusion: the work that decides whether AI pays off sits in the data platform. The products Databricks highlighted map onto those needs. Lakebase covers live operational data, Genie supplies business context, and the gateway addresses the cost sprawl that comes from running many models.

The growth figures suggest buyers agree: a warehousing business growing faster than 100% a year points to enterprises consolidating analytics and AI workloads on platforms they expect to serve agents.

What are the reasons for caution?

The first is valuation. By the magazine's own calculation, not a figure Databricks or its investors disclosed, $190 billion against a run-rate above $7 billion works out to roughly 27 times annualized revenue, and somewhat less if the run-rate is well above $7 billion. The same arithmetic puts the December round at about 28 times, so the multiple has held roughly steady while revenue grew. Run-rate is a forward-looking snapshot, not audited annual revenue, and as a private company Databricks chooses which metrics to disclose. The release, for example, did not give a net revenue retention figure or growth rates for Agent Bricks or Genie.

The second is concentration. Fortune reported on August 12, in a story drawing on PitchBook's second-quarter valuation data, that AI megadeals took 87.5% of U.S. venture dollars in the first half of 2026. "The winners are bigger than ever, overshadowing the rest of the venture market," PitchBook senior research analyst Emily Zheng told Fortune. That favors scale players like Databricks, but it also means expectations are priced in, and the companies on the other side of the table may feel pressure to show returns quickly.

The third concern belongs to customers: lock-in. A platform that bundles database, warehouse, context and model gateway is convenient, but it raises the cost of leaving. AI Connect speakers have pushed for openness as a hedge. At an AI Demo Day webinar on August 5, Kelly Kohlheffer, Senior Global Director, Partner Sales Engineering at Fivetran, showed one dataset in open formats queried by Snowflake, Databricks, Athena, BigQuery, DuckDB, Microsoft Fabric and ClickHouse with matching answers, summing it up as "seven engines, five catalogs, three open formats, one copy." AtScale CTO Dave Mariani made the strategic case in a webinar on AI infrastructure at scale, warning against semantic layers tied to one platform because "we don't know where this stuff is going."

What does this mean for enterprise AI and data leaders?

Budget for the foundation, not just the model. Databricks' numbers are evidence that large enterprises are already moving AI spend into data infrastructure. Leaders still funding AI mainly as model licenses and pilots may be underinvesting in the layer that determines whether agents can be trusted.

Treat cost control as architecture. The prominence of Unity AI Gateway in the release reflects a real problem: multi-model AI bills are hard to predict. Whatever platform you use, put routing, quotas and per-team attribution in place before agent usage scales.

Keep exit options open. Open table formats and portable semantic definitions reduce dependence on any single vendor, as the Fivetran demo showed. That matters more as platforms add more of the stack.

Use the vendor's momentum in negotiations, carefully. A fast-growing supplier with fresh capital has room to invest in your success, but also every incentive to expand its footprint. Tie commitments to measured outcomes, not consumption alone.

The next markers to watch are whether Databricks discloses retention and AI product revenue as it scales, and how rivals respond on price and openness. Whatever happens to the valuation, the round confirms a point AI Connect practitioners have made repeatedly: enterprise AI value is being decided in the data layer.

Frequently asked questions

How much did Databricks raise in August 2026?

Databricks raised $5 billion at a $190 billion valuation in a round led by Coatue, announced August 13, 2026.

What is Databricks' revenue run-rate?

Databricks said its revenue run-rate passed $7 billion, with more than 80% year-over-year growth in its second quarter and positive adjusted free cash flow over the last 12 months.

What should enterprises take from the Databricks round?

That AI budgets are shifting toward data infrastructure, context and cost controls, and that leaders should pair that investment with open formats and outcome-based vendor commitments.

Filed under: Data & AI Infrastructure, AI Strategy & ROI, Context, Semantics & RAG