Databricks hits $190bn valuation as Lakebase and AI agent strategy expand

Databricks Inc. has closed a $5 billion strategic funding round at a $190 billion valuation after reporting a revenue run-rate above $7 billion and year-over-year growth exceeding 80% in its fiscal second quarter. Coatue led the financing, with Blackstone, MGX, accounts advised by T. Rowe Price and new investor Sixth Street Growth among the major participants, while a broad group of new and existing institutional investors also joined the round. Databricks plans to direct the capital toward products including Lakebase, Genie and Unity AI Gateway as it tries to become the underlying data, database and governance layer for enterprise AI agents. The latest valuation is approximately 42% above the $134 billion level attached to Databricks’ February financing, while the company continues to report positive adjusted free cash flow over the trailing 12 months. The central tension is no longer whether Databricks can grow rapidly, but whether a private company already valued well above many listed software groups can expand far enough beyond analytics and data warehousing to justify a $190 billion price before an eventual IPO.

How did Databricks move from a $134 billion valuation to $190 billion in roughly six months?

The progression between Databricks’ two 2026 financing events is unusually large even within the current artificial intelligence investment cycle. In February, the company announced approximately $5 billion of equity financing at a $134 billion valuation alongside roughly $2 billion of additional debt capacity. At that stage Databricks had surpassed a $5.4 billion revenue run-rate, was growing more than 65% year over year and had remained free-cash-flow positive over the preceding 12 months.

By August 13, the disclosed revenue run-rate had moved above $7 billion and quarterly year-over-year growth had accelerated beyond 80%. The valuation increased by $56 billion, or approximately 41.8%, between the two reference points, while the minimum disclosed revenue run-rate increased by at least 29.6%. That means investors have rewarded not only the additional revenue but also the expectation that Databricks’ strategic position within enterprise artificial intelligence has become more valuable.

A simple valuation-to-run-rate comparison illustrates the point. The February valuation represented roughly 24.8 times the $5.4 billion revenue run-rate. Using exactly $7 billion as a conservative denominator, the new valuation would represent about 27.1 times revenue run-rate. Because Databricks says revenue has surpassed $7 billion, the actual multiple is somewhat below that figure, but the calculation still indicates that valuation appreciation has broadly kept pace with, and may have outstripped, reported commercial expansion.

The higher valuation also finalises a process that began in July, when Databricks signed a term sheet for strategic financing at a $188 billion valuation led by Coatue. The final round closed at $190 billion and expanded to $5 billion, with additional investors joining the transaction. Databricks therefore secured a slightly higher final valuation than the July reference level rather than simply closing the original transaction on unchanged terms.

The investor roster demonstrates the depth of private capital available to Databricks. New participants included BOND, Clearlake Capital, Point72, Premji Invest and TPG, while existing investors included Andreessen Horowitz, Fidelity Management & Research Company, Franklin Templeton, GIC, Goldman Sachs Alternatives, Insight Partners, J.P. Morgan Private Capital, Morgan Stanley Investment Management, Ontario Teachers’ Pension Plan, Temasek and Thrive Capital. The mix extends well beyond traditional venture capital into private equity, sovereign capital, pension funds and major asset managers.

That depth gives Databricks an unusual strategic option. Unlike many venture-backed software businesses, it does not appear to need an immediate public listing simply to fund operations. Management can continue investing privately while waiting for an IPO environment and valuation framework it considers more favourable.

Why does the $7 billion revenue run-rate make this more than another AI funding story?

Databricks’ revenue scale differentiates the transaction from funding rounds built primarily around future market potential. The company says it has exceeded a $7 billion revenue run-rate, with growth above 80% year over year, while adjusted free cash flow has remained positive over the past 12 months. It now reports more than 1,000 customers consuming its services at an annual revenue run-rate above $1 million and more than 100 consuming at a run-rate above $10 million.

Those customer metrics have expanded materially since February. Databricks then reported more than 800 customers above the $1 million annual run-rate threshold and more than 70 above $10 million. In approximately six months, the count of million-dollar customers has therefore increased by at least 25%, while the number of customers above $10 million has increased by more than 42%.

Large customer expansion is especially important for a consumption-oriented cloud software business. Databricks does not need to acquire a completely new customer for every incremental dollar of revenue. Existing organisations can increase usage as they move additional data engineering, warehousing, machine learning and artificial intelligence workloads onto the platform.

Earlier disclosures reinforce that mechanism. In February, Databricks reported net retention above 140%, meaning the comparable customer cohort was expanding spending substantially even after accounting for contractions. The August announcement does not provide an updated net-retention figure, so the earlier metric should not be assumed unchanged, but it helps explain how Databricks has been able to expand revenue more rapidly than its underlying enterprise customer count.

The company now says more than 20,000 organisations use Databricks, including approximately 70% of the Fortune 500. That moves the growth problem from establishing enterprise credibility toward capturing a larger share of technology spending inside organisations that already use the platform.

This is strategically valuable because large enterprises rarely replace their entire data stack in one transaction. Databricks can instead expand across adjacent workloads. A customer may begin with data engineering, add warehousing, introduce machine learning, deploy AI agents and eventually use Databricks for operational databases, governance and employee-facing business intelligence.

The $7 billion revenue run-rate is therefore important not simply because of its size. It provides evidence that Databricks already owns enough enterprise workloads for adjacent products to have an existing distribution channel.

Why are Lakebase, Genie and Unity AI Gateway central to the $190 billion valuation case?

Databricks is directing the new capital toward three products that illustrate how far it wants to move beyond its original data analytics position. Lakebase provides a serverless Postgres database for applications and AI agents. Genie allows employees and agents to work with enterprise information using natural language. Unity AI Gateway provides governance, routing and cost controls across multiple artificial intelligence models.

Lakebase may be the clearest evidence that the expansion is already becoming commercial. Databricks says the product has exceeded a $100 million revenue run-rate despite being relatively new. Its existing Lakehouse data warehousing product has surpassed a $1.5 billion revenue run-rate and is growing more than 100% year over year.

Lakebase grew out of Databricks’ acquisition of Neon, the serverless Postgres company it agreed to buy in 2025. Databricks had observed that more than 80% of databases created on Neon’s system were being provisioned automatically by AI agents rather than human developers, providing an early indication that autonomous software could create database demand very differently from traditional application development.

That pattern matters because agents require persistent operational information rather than merely access to a large analytical repository. An agent handling customer service, pricing, inventory or financial workflows may need to read current information, write new records, retain state and perform actions. Databricks is attempting to connect those operational requirements with the analytical and governance layers it already supplies.

Genie addresses a different bottleneck. Large enterprises possess extensive data but frequently struggle to make it accessible to employees without specialised analytical skills. Databricks is positioning Genie as an AI coworker capable of turning business data into answers and actions, potentially allowing the company to move closer to the end user rather than remaining an infrastructure product primarily accessed by technical teams.

Unity AI Gateway addresses another emerging problem: companies are unlikely to rely on a single artificial intelligence model indefinitely. Enterprises may use models from OpenAI, Anthropic, Google and other providers depending on workload, price, privacy and performance. A governance layer that controls model access, monitors expenditure and routes workloads among providers could become increasingly important as agent deployments multiply.

The strategic logic is straightforward. Databricks already sits close to enterprise data. If it can also control the databases agents use, the context they receive and the models they call, the company can occupy a much larger portion of the AI application stack.

The execution risk is equally clear. Every additional layer brings Databricks into competition with another category of software provider.

How does Databricks’ $190 billion valuation compare with publicly traded rival Snowflake?

The most direct public-market comparison remains Snowflake Inc. (NYSE: SNOW), which competes with Databricks across cloud data platforms, analytics and increasingly enterprise artificial intelligence. Snowflake shares were trading around $334.73 on August 13, giving the company a market capitalisation of approximately $115.6 billion. Databricks’ $190 billion private valuation is therefore about 64% higher than Snowflake’s current public equity value.

The difference partly reflects growth. Snowflake reported fiscal first-quarter 2027 product revenue of $1.33 billion, representing 34% year-over-year growth, and raised its full-year product revenue guidance to $5.84 billion. Databricks is reporting a company-wide revenue run-rate above $7 billion and growth above 80%. The measures are not directly identical because Snowflake reports recognised product revenue while Databricks is disclosing an annualised revenue run-rate, but the gap in reported growth rates is substantial.

Snowflake nevertheless offers a useful reminder of what changes once a company is publicly traded. Public investors can evaluate quarterly revenue, margins, stock-based compensation, free cash flow, guidance and remaining performance obligations in detail. Databricks releases selected operating indicators, but outsiders still lack the full financial statements necessary to analyse gross margin, operating expenses and earnings quality with the same precision.

Public markets can also reprice software valuations quickly. Snowflake’s current market value reflects daily trading against changing expectations for growth, competition and artificial intelligence adoption. Databricks’ $190 billion mark is established through a negotiated private financing involving sophisticated investors but does not face continuous price discovery.

The comparison therefore cuts both ways. Databricks’ higher growth helps explain why private investors are willing to place a higher value on the company. Staying private also means that valuation does not yet undergo the same recurring earnings tests applied to Snowflake.

The next Databricks financing may matter less than its eventual IPO precisely because a listing would reveal how public investors price a business growing far faster than most established software companies while already carrying one of the technology industry’s largest private valuations.

Can Databricks use AI agents to move beyond data infrastructure into traditional enterprise software?

The most strategically ambitious part of Databricks’ current expansion is its movement closer to applications and business workflows. AI agents need access to corporate information to make useful decisions, creating an opportunity for a company that already manages large quantities of enterprise data.

This could weaken one of the traditional advantages enjoyed by application-software companies. Salesforce, Adobe and other enterprise vendors have historically benefited from owning the applications where customer, marketing and operational information is created and stored. Databricks is approaching the problem from the opposite direction: control the data and intelligence layer first, then allow agents and applications to operate directly on top of it.

The strategy became more explicit with CustomerLake, Databricks’ agentic customer-data platform introduced in June. The product combines customer information, analytics and AI capabilities in an attempt to support marketing and customer workflows from the same underlying data environment. Reuters Breakingviews argued in July that this direction could allow Databricks to challenge parts of the traditional application-software market rather than remaining confined to data infrastructure.

That represents potentially enormous upside but also creates a broader competitive field. Databricks already competes with Snowflake in data infrastructure. Lakebase places it closer to database providers. Genie overlaps with business intelligence and enterprise assistant products. CustomerLake increases competition with marketing technology providers. Agent Bricks brings it into the rapidly expanding market for enterprise AI agents.

Microsoft, Alphabet and Amazon Web Services create another dimension because each company provides cloud infrastructure, databases, AI models and application tooling. Databricks maintains significant partnerships with the major cloud platforms rather than competing with them in every category, but dependence and competition increasingly coexist as both sides expand their product portfolios.

Databricks recently expanded its Microsoft partnership into the 2030s, including greater use of Microsoft Azure and Azure Cobalt processors and deeper product integration. The arrangement illustrates the company’s unusual position. It is large enough to compete for enterprise workloads while remaining one of the largest consumers and partners of the hyperscale cloud providers.

The strategic objective is not necessarily to displace every application vendor. Databricks can justify substantial growth simply by becoming the governed platform through which corporate agents access information and execute tasks across existing applications. In that scenario, the company benefits even when Salesforce, Microsoft or another vendor remains the system of record.

Why does another $5 billion private round make a near-term Databricks IPO less necessary?

Databricks has been viewed as a major IPO candidate for years, but the company’s financial position reduces the urgency to list. The February financing already combined approximately $5 billion in equity with about $2 billion of debt capacity while Databricks was generating positive free cash flow. The new $5 billion strategic round further increases the capital available for research, product development, acquisitions and employee liquidity.

Chief Executive Officer Ali Ghodsi indicated earlier in 2026 that Databricks continued to prepare for an eventual public listing but was willing to remain private while market conditions favoured private financing. The ability to raise billions of dollars from institutional investors allows the company to separate access to capital from the decision to list.

That is increasingly uncommon at this scale. A conventional startup eventually reaches a point where public markets are required to provide sufficient growth capital and liquidity. Databricks has instead assembled a private shareholder base containing many of the same institutions that would potentially buy shares in an IPO.

Remaining private offers several advantages. Management can invest aggressively without quarterly stock-price pressure, negotiate employee liquidity programmes privately and time an IPO around strategic rather than immediate financing requirements. It can also continue using a high private valuation as currency when pursuing acquisitions.

There are costs. A growing employee base eventually needs liquid equity, private financing structures can become increasingly complex and investors may demand paths to realisation after committing capital across successive rounds. The larger Databricks becomes, the more unusual indefinite private ownership would be.

The latest round therefore delays rather than eliminates the IPO question. In fact, a $190 billion valuation raises the eventual listing threshold. Databricks would probably want public investors to support a valuation at least comparable to the latest private financing, making sustained revenue growth and cash generation critical in the quarters before any offering.

What are the key takeaways from Databricks’ $5 billion funding round and $190 billion valuation?

• Databricks closed $5 billion of strategic financing at a $190 billion valuation on August 13, 2026.
• Coatue led the round, with Blackstone, MGX, T. Rowe Price accounts and Sixth Street Growth among the major participants.
• The valuation is approximately 41.8% above the $134 billion level announced in February.
• Databricks has surpassed a $7 billion revenue run-rate with year-over-year growth exceeding 80%.
• The company continues to report positive adjusted free cash flow over the trailing 12 months.
• More than 1,000 customers now consume Databricks at annual revenue run-rates above $1 million, with more than 100 above $10 million.
• Lakebase has already surpassed a $100 million revenue run-rate, while the Lakehouse warehousing business has crossed $1.5 billion.
• Databricks’ $190 billion private valuation is roughly 64% above Snowflake’s approximately $115.6 billion public-market capitalisation on August 13.
• The funding reduces immediate pressure for an IPO and gives Databricks substantial capital for AI products and potential acquisitions.
• Continued revenue growth, cash generation and successful expansion beyond data infrastructure will determine whether the $190 billion valuation can survive eventual public-market scrutiny.

What will determine whether Databricks can ultimately grow into a $190 billion valuation?

Databricks enters this financing from a stronger operating position than most private companies receiving large artificial intelligence valuations. Revenue has moved above a $7 billion annualised run-rate, growth has accelerated beyond 80%, the company remains adjusted-free-cash-flow positive and the number of customers spending at seven-figure and eight-figure annual rates continues to increase. These are substantial operating indicators rather than a valuation built solely around future artificial intelligence potential.

What has become more demanding is the price attached to those achievements. Databricks’ valuation has increased by $56 billion since February, and its private equity value now stands materially above publicly traded Snowflake. The company therefore needs to sustain exceptional growth for long enough that revenue catches up with the valuation rather than relying on continued multiple expansion.

The strongest proof would come from Lakebase, Genie, Unity AI Gateway and related products becoming material revenue contributors without weakening free cash flow. Lakebase crossing $100 million already provides an early signal, but a $190 billion valuation requires several large product franchises rather than one successful adjacent launch.

The thesis would strengthen if Databricks maintains high enterprise expansion rates, converts its installed data-platform base into agent and database revenue and preserves positive cash generation while investing aggressively. It would weaken if growth normalises before new products become large enough to compensate, if hyperscalers or Snowflake narrow the AI product gap, or if Databricks must spend disproportionately more to defend each additional layer of the platform.

The $5 billion financing gives Databricks considerable freedom to postpone its public-market test. It does not remove that test. At $190 billion, the eventual question for investors will be much larger than whether Databricks can beat Snowflake in cloud data. Databricks must demonstrate that controlling enterprise data gives it a defensible route into the database, agent and application layers that will determine where corporate AI spending ultimately settles.