Data lakehouse business Databricks is said to be in talks to raise $5 billion, just three months after closing a $1 billion K-round of VC funding.
This would take Databricks total funding to over $20 billion, a testament to the perceived worth of the company’s lakehouse technology, which aggregates widely-scattered data and makes it available for analysis. Particularly these days, it can be used to train AI models and agents, and in AI inferencing.
Databricks issued a press release in September saying it was closing a $1 billion K-round of funding at at greater than $100 billion valuation. At the time, it said it had crossed a $4 billion revenue run rate during its second calendar quarter, from 15,000 customers, was growing in excess of 50 percent Y/Y, and was cash flow positive. (In June this year, Databricks expected its annual run rate revenues would be $3.7 billion.) Its annual revenue growth rate is high, as Databricks pulled in $2.6 billion in revenues in its 2024 year, and $1.3 billion in 2023.
Now, three months after the K round, various financial media outlets are saying it’s discussing a $5 billion raise at a $134 billion valuation, with a $4.1 billion revenue run rate; not that much larger than the September guidance.
The valuation behind this potential $5 billion funding round is said to be $134 billion, which is a 32.7 multiple of its projected full year revenues. Last December, when it raised $10 billion, the valuation was $62 billion. These valuation numbers seem like fantasy figures.
For comparison, publicly-owned Snowflake is capitalized at $89.59 billion. It reported $1.09 billion in revenues from its 12,062 customers for its second fy2026 quarter, ended July 31, up 32 percent Y/Y; a lower revenue growth rate than Databricks. That gives it a $4.36 billion annual revenue run rate, and it’s guiding $4.4 billion in full fy2026 revenues. Its capitalization values the company at 20.36 times its annual revenues.
Databricks has a net dollar retention rate of 140 percent, and its gross margin is said to be 74 percent, down from 77 percent earlier in the year. This is attributed to high customer take up of its AI facilities, lifting its compute costs. With its 327 valuation multiple investors must be hoping to see continued high growth with a knockout IPO next year.
A Time magazine article quotes Databricks’ CEO Ali Ghodsi as believing AGI (Artificial General Intelligence) has already been achieved, and saying:“Now that we have it, it’s kind of like, never meet your heroes. It’s just like, okay, that’s not that impressive.”
Apart from that, Databricks is helping its customers train task-focussed AI agents with its Agent Bricks tech. These should cost less to run than ChatGPT or Claude and do better at their specific tasks. If it achieves that then a future IPO could indeed be a knockout.
