Databricks Eyes $175B Valuation on $5.4B ARR — The Data Layer Is Now Worth More Than Most Foundation Model Labs
Databricks is in discussions to raise a new funding round at a valuation of $165 billion to $175 billion, according to The Information. The round could begin as early as next month. The company surpassed a $5.4 billion revenue run rate in February 2026, up 65% year over year, with AI products alone generating $1.4 billion in annualised revenue.
The reporting follows Databricks’ February round: $7 billion at $134 billion, split $3 billion equity and $2 billion debt, led by JPMorgan, with Goldman Sachs, Morgan Stanley, Neuberger Berman, and the Qatar Investment Authority participating. That round closed at $100 billion in August 2025. In four months it was $134 billion. Now discussions point to $175 billion.
What the Valuation Reflects
At $175 billion on $5.4 billion ARR, the implied multiple is approximately 32x forward revenue. That is expensive but not irrational for a company growing at 65%. For context: at the same growth rate over two years, Databricks crosses $10 billion in ARR before most AI foundation model labs reach their first profitable quarter.
The $1.4 billion in AI product revenue is the more telling number. It separates Databricks from pure-play cloud databases: the company is generating real revenue from the AI layer, not just hosting the data pipelines that AI systems run on. Enterprise customers are paying for Databricks-native AI capabilities, not just storage and compute.
The Infrastructure Thesis
Foundation model companies attract the headlines. Databricks makes the case that data infrastructure underneath those models is the durable business. Its argument: every enterprise AI deployment needs data management, governance, and quality tooling regardless of which foundation model wins the benchmark war. Customers do not reprovision Databricks when they switch from Claude to GPT-5.5. They reprovision their model endpoint.
CEO Ali Ghodsi has told investors the company is IPO-bound, potentially as early as 2027. He called 2026 the “worst year” to go public given the crowded listing calendar: SpaceX priced at $135 on June 12, Anthropic filed its S-1 at $965 billion, and OpenAI is targeting a $1 trillion debut. Databricks can afford to wait; its growth rate generates negotiating leverage over underwriters.
The Risk
The hyperscaler competition is the structural threat no valuation analysis can ignore. AWS, Azure, and Google Cloud all offer overlapping data lake and AI analytics products. Databricks competes with its largest customers’ compute infrastructure businesses simultaneously. At $175 billion, investors are betting that Databricks’ data governance and AI product layer is sufficiently differentiated to hold margin as hyperscaler products improve.
That is not a small bet. But at 65% growth with $1.4 billion in AI product revenue already, it is a bet with a working engine behind it.