Databricks hits $188B valuation, extending its run as AI’s favorite second act
TechCrunch Julie Bort
Databricks just landed a $188B valuation in a fresh funding round led by Coatue. That's up from $134B in February — its fourth mega-round in under two years.
Based on reporting by TechCrunch, Julie Bort — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Databricks announced Thursday that it's now valued at $188 billion, in a round led by Coatue that hasn't technically closed yet. The company says the cash isn't in hand and won't be until later this summer, which is a strange thing to broadcast before the wire transfer clears. But a venture capitalist told TechCrunch the deal is real and oversubscribed enough that Databricks had zero reason to sit on the number quietly.
This is not a one-off spike. Databricks raised $5 billion at $134 billion back in February. Before that, $1 billion at $100 billion in September 2025. Before that, a then-record $10 billion round at $62 billion in December 2024. The pace has been so relentless that people online started joking Databricks would run out of letters for its Series naming scheme, with one person quipping they'd turned on alerts for the eventual Series AA.
What's actually driving this is a genuine reinvention. Databricks built its name in the big-data era, selling enterprises on cloud storage paired with fast analytics. That history turned out to be an accidental advantage once companies started demanding AI tools with the same governance and security guarantees they already expected from their data stack. Databricks answered with a wave of AI-specific products — Lakebase for agent-facing databases, Unity as an AI gateway, and a coordination layer called Omnigent for managing swarms of agents.
The company has also become a poster child for a trend that's defining 2026: enterprises quietly swapping expensive proprietary models for cheaper Chinese open-weight ones. CEO Ali Ghodsi published internal benchmarks last week showing Z.ai's GLM 5.2 now handles even the hardest coding tasks his 3,000 engineers throw at it, and does so more cheaply than anything from OpenAI or Anthropic. The twist nobody expected was that the harness — the wrapper software managing context and prompts around the model — mattered just as much as the model itself, with an open-source tool called Pi outperforming pricier alternatives on cost without losing quality.
None of this means Databricks is an AI lab in the way Anthropic or OpenAI are. It's a data company that read the room early and pivoted hard enough to earn AI's halo effect, the same glow that's now so bright even Jersey Mike's mentioned AI 22 times in an S-1 filing. That halo happens to be worth billions in valuation, and investors clearly aren't done paying for it.
My take — AI-written commentary, not fact-checked reporting
I'll believe the $188 billion when the check clears, but the bigger story here isn't the number — it's that Databricks proved you can out-earn your own hype by actually building infrastructure people need, instead of just slapping 'agentic' on a slide deck. The GLM 5.2 detail is the real signal: enterprises are voting with their compute budgets for open-weight models over premium American ones, and that's a trend the Anthropic-OpenAI duopoly should be sweating over, not the valuation headline.
Read more about this at: TechCrunch