Databricks is reportedly in talks for a $165-175 billion valuation as of June 2026, up from $134 billion just four months earlier, after annualized revenue hit $6.9 billion growing 80% year-over-year.
That's the short answer. The longer answer is that Databricks is now worth roughly double Snowflake's public market cap despite similar revenue, has stayed free-cash-flow positive while doing it, and keeps pushing its long-rumored IPO further into the future even as every metric points toward one.
Databricks Valuation 2026: The Complete Breakdown
Databricks' valuation in 2026 has moved in two fast steps: a $134 billion mark set in February on a $5 billion equity raise (plus $2 billion in added debt capacity), followed by reports in June that the company is in talks for a new round targeting $165-175 billion. The jump tracks almost exactly with revenue acceleration โ annualized revenue grew from $5.4 billion in February to $6.9 billion by June, with year-over-year growth climbing from 65% to roughly 80% over the same stretch.
For context on how private markets price AI-era software companies more broadly, see our breakdown of the 10 highest-valued private AI companies in 2026 and track live comps on the AI Valuations dashboard.
Founded in 2013 by the creators of the Apache Spark open-source project, Databricks spent its first decade positioned as infrastructure for data engineers โ the pipes that move and process enterprise data before anyone builds an application on top of it. That positioning is exactly why the 2026 valuation surprises people who haven't followed the company closely: a business once thought of as unglamorous plumbing has become one of the largest AI infrastructure bets in venture history, second only to the foundation-model labs themselves in private-market size.
Databricks Annualized Revenue Run-Rate, 2025-2026
Revenue growth accelerated from 55% to 80% YoY even as the ARR base nearly doubled
Databricks company press releases, 2025-2026.
Databricks vs Snowflake: Comparing Valuation, Revenue, and Growth
The natural comparison for Databricks' valuation is Snowflake, the public data-cloud company it competes with head-to-head for enterprise data and AI workloads. On revenue the two are close. On valuation, Databricks is worth far more โ the market is pricing Databricks' growth rate, not just its current revenue.
| Metric | Databricks | Snowflake |
|---|---|---|
| Valuation / market cap | $134B (Feb), talks at $165-175B (Jun) | ~$83B public market cap |
| Annualized revenue | $6.9B (Jun 2026) | ~$5.6B (roughly $4.68B FY2026 total revenue) |
| YoY growth rate | ~80% (Q2 2026) | 27-29% guided product growth |
| AI product revenue | $1.4B run rate (~26% of ARR) | AI strategy still pivoting per reporting |
| Net dollar retention | 140%+ | ~125% (historically disclosed range) |
| Cash flow status | FCF-positive for full-year 2025 | Public, GAAP profitable on a non-GAAP basis |
| Public listing status | Private; IPO targeted late 2026 or 2027 | Public since 2020 IPO |
Figures are 2026 estimates blended from Databricks press releases, CNBC, The Information, PYMNTS, and Snowflake public filings. Databricks metrics are self-reported run-rate figures, not audited GAAP revenue.
Databricks vs Snowflake: Valuation vs Annualized Revenue ($B)
Databricks trades at roughly 2x Snowflake's valuation on similar revenue, reflecting its 3x faster growth rate
The Information, CNBC, PYMNTS, company disclosures, June 2026.
What's Actually Driving Databricks' 2026 Valuation Higher
Three things are pushing Databricks' valuation up faster than almost any other private software company: accelerating growth at massive scale, a real AI product line rather than a bolted-on feature, and capital discipline that lets it raise without burning through the proceeds.
Growth is re-accelerating, not decaying
80% YoY growth in Q2 2026 is up from 65% in Q4 2025 โ unusual at a $6.9B revenue base, where growth almost always slows
AI products are real revenue, not a narrative
$1.4B in AI product run rate is roughly 26% of total ARR, driven by Mosaic AI and Databricks-native model tooling
Data Warehousing is compounding fast
That segment alone crossed $1.5B ARR by June 2026, up from $1.0B in Q3 2025 โ a 50% jump in nine months
It doesn't need the cash
FCF-positive for all of 2025 means every dollar raised goes to growth and buybacks for early employees, not survival
Net dollar retention above 140% is the number that matters most to the investors writing the $165-175B check. It means existing customers are increasing their Databricks spend by more than 40% a year on average, independent of new-logo growth โ the same dynamic that let Snowflake command a premium multiple in its own early years as a public company, before its growth decelerated to the high-20s. Compare how multiples move with ARR scale on our SaaS valuation multiples by ARR range post.
The competitive backdrop matters too. Databricks isn't just outrunning Snowflake โ it's also fending off Google BigQuery, Microsoft Fabric, and Amazon Redshift, each backed by a hyperscaler with effectively unlimited capital and a captive cloud customer base. That Databricks keeps taking share from vendors with that kind of distribution advantage is itself a data point investors weigh heavily: it suggests the product, not the sales channel, is winning the deal.
Who Is Actually Backing the Databricks Valuation
The February 2026 round wasn't a niche syndicate โ it was led by Insight Partners, Fidelity Management & Research, and J.P. Morgan Asset Management, with Andreessen Horowitz, BlackRock, Blackstone, Coatue, GIC, MGX, NEA, Ontario Teachers' Pension Plan, Robinhood Ventures, T. Rowe Price, Temasek, Thrive Capital, and Winslow Capital all participating. That roster looks more like a pre-IPO crossover book than a typical late-stage venture round, and it's exactly the kind of investor base that shows up in the final 12-18 months before a listing.
The demand side of that valuation is just as concrete. Databricks counts more than 20,000 customers, including adidas, AT&T, Bayer, Block, Mastercard, Rivian, and Unilever, and says over 60% of the Fortune 500 now run on its platform. More tellingly, over 700 of those customers each generate more than $1 million in annual revenue run-rate for Databricks โ the cohort that drives the 140%+ net dollar retention and gives investors confidence the $6.9B ARR base keeps compounding without heroic new-logo growth. Total funding raised to date sits at roughly $20.2 billion across 14 rounds and 110 investors, making the February round alone worth more than a third of everything raised in the company's history.
Will Databricks IPO Before the Next Valuation Reset?
Databricks CEO Ali Ghodsi has said the company is in "IPO preparation mode," with a listing possible as soon as late 2026 โ but he also told Bloomberg Television in June 2026 that this is "a terrible year" to go public, citing competition for investor attention from marquee listings like SpaceX. That combination of readiness and reluctance has defined Databricks' IPO story for three years running.
The practical effect is that Databricks keeps raising private capital instead โ $5B in February 2026, talks of a new round in June โ rather than testing public markets. Every private round at a higher price makes the eventual IPO a bigger, higher-stakes event, and raises the bar for what counts as a successful debut. Track how the broader IPO pipeline is shaping up on our Tech IPO tracker.
Databricks added roughly $30-40 billion in valuation in four months without going public.
$165-175B on the table, $6.9B in revenue growing 80% a year, and an IPO that keeps getting rescheduled.
At some point the private markets run out of room to keep re-pricing a company this size. Until then, Databricks is proof that staying private longer is now a valuation strategy, not just a delay tactic.
Track live private-company valuations on the AI Valuations Dashboard and SaaS Valuations Dashboard at Value Add VC. Originally published in the Trace Cohen newsletter.
Get VC data most people never see โ free.
Weekly benchmarks, valuations, and fund data. No spam, unsubscribe anytime.