SAP completed its Prior Labs acquisition on Friday, 17 July 2026, taking full control of the German AI startup behind tabular foundation models and pledging more than €1 billion over four years to scale it into a frontier lab.
Freiburg-based Prior Labs pioneered tabular foundation models, a type of AI that reads spreadsheets and structured business data the way large language models read text.
As reported by SAP News Centre, the completed deal follows the intention SAP first announced in May 2026, and Prior Labs will keep operating as an independent entity inside the company.
What SAP’s Prior Labs acquisition actually buys
A tabular foundation model, or TFM, is the nerdy term for an AI trained to make sense of the rows and columns that businesses live in, from sales ledgers to sensor logs. R
ather than predicting the next word, it predicts the next value, spotting the patterns hiding inside a spreadsheet.
Most of the AI industry poured its money into large language models like ChatGPT. Prior Labs bet on the less glamorous prize, the structured data that runs the world’s businesses, an area where language models struggle.
SAP committed more than €1 billion over the next four years to chase it.
Prior Labs co-founder Frank Hutter framed the deal as a step change.
“Eighteen months ago, Prior Labs was a research project. Today, we’re beginning our next chapter as an AI lab with the resources to tackle problems we simply couldn’t before. Taking tabular foundation models to the next level requires better data environments, deployment surfaces, and long-term research investment, and SAP is uniquely positioned to provide all of these.”
How Prior Labs built the tabular foundation model
Prior Labs was founded in 2024, roughly 18 months before the deal closed, by Freiburg professor and automated-machine-learning expert Frank Hutter, computer scientist Noah Hollmann, a Google and Boston Consulting Group alumnus, and Sauraj Gambhir, who brought a venture capital and M&A background.
It had raised only €9 million in a single pre-seed round.
The team created TabPFN, the foundation model that opened the field. The research reached the journal Nature and has been downloaded roughly 4 million times.
It already works in the wild, with Hitachi predicting train failures, TD using it for financial forecasting, and researchers applying it across pancreatic cancer diagnosis and wildfire prediction.
SAP’s chief technology officer Philipp Herzig cast the bet as an early read on where enterprise AI was heading.
“Early on, SAP recognised that the greatest untapped opportunity in enterprise AI wasn’t large language models; it was AI built for the structured data that runs the world’s businesses. Prior Labs has defined the category of TFMs and has built the world’s strongest research team in this category, topping the public benchmarks since day one.”
What the Prior Labs acquisition means next
Over the next four years, SAP will keep funding Prior Labs as a standalone lab, letting it hold on to its brand, leadership and research focus.
The company said the startup would carry on publishing its work openly, keeping the door open for the outside researchers who pushed TabPFN this far.
