Mistral AI announces a partnership with Cloudera, aimed at integrating its language models into Cloudera’s hybrid data platform.
The stated objective is to meet the demand for “sovereign AI” in regulated sectors (finance, industry, telecom), while keeping “data, intelligence, compute and operations under the client’s control,” according to Mistral’s official post.
The partnership rests on two main axes.
The Mistral models will be integrated into Cloudera’s platform, enabling enterprises to deploy them “in private and public cloud environments, on-premises and fully isolated (air-gapped), while maintaining full control,” Mistral notes.
Additionally, Mistral offers companies the ability to train their models “on large quantities of proprietary data in controlled environments,” while preserving “ownership of both the data and the resulting intelligence.”
Concretely, Cloudera’s customers will be able to run inference, generation and agent workflows directly without sending their data to external APIs.
« From Renting Generic AI to Owning Intelligence »
For Cloudera, the aim is to move companies “from renting generic AI to owning intelligence that is theirs,” explains Abhas Ricky, Chief Business Officer and GM Applied AI at Cloudera.
“General-purpose models are the starting point, not the finish line. The real advantage comes from models trained on decades of proprietary data: credit decisions, production cycles, network telemetry that no one else possesses,” he adds.
“It’s a privilege to have the opportunity to bring Mistral’s sovereign AI to the 30 exabytes of data managed by Cloudera’s customers who run on its platform,” emphasizes Kamal Brar, SVP Partnerships & Alliances.
The two companies have not disclosed the exact figure for this deal.
Several trade outlets, however, report a “nine-figure strategic partnership” (i.e., between $100 million and $999 million), with no further specifics.