French startup Mistral AI just landed €3 billion to build its own computing backbone, pushing its valuation past €21 billion and cementing its role as Europe’s sovereign answer to American AI dominance.
The three-year-old company completed what it calls the largest equity raise by any European tech firm. Samsung Electronics led the Series D round, joined by EQT’s Scaleup Europe Fund and returning investor PSG Equity. New backers include Advent, BlackRock-managed funds, and Luxembourg’s government, alongside a roster of prior investors.
**Why Europe needs its own AI player**
Governments and corporations increasingly want control over where AI systems run, where data lives, and who holds the keys to infrastructure. Mistral has made that anxiety central to its pitch.
Its open-weight models let customers download, modify, and deploy AI inside their own environments rather than depending on rented access from US providers. That flexibility matters for enterprises protecting intellectual property and for European governments wary of foreign tech dependence.
“The fact that the EU or Europe have to have their own kind of AI provider in the game is important,” CFO Johan Bergqvist said.
More than 125 global companies now use Mistral’s technology, including Airbus, ASML, and HSBC.
**Building compute, not just models**
The fresh capital will accelerate a shift toward owned infrastructure. CEO Arthur Mensch said the company’s self-operated compute capacity should roughly double within five years as it reduces reliance on rented resources.
“Long term, the plan is to fully rely on capacity that we are building ourselves,” Mensch said.
Mistral already has a data centre project planned in Sweden. The company projects annual recurring revenue will top $1 billion this year.
**Open source as leverage**
While OpenAI and Anthropic guard their proprietary systems, Mistral sells openness. Bergqvist said open source resonates with corporations that want sovereignty over their AI transformation and tighter control of customer data and IP.
That approach also shields buyers from a single vendor’s pricing shifts or roadmap changes.
Competition remains fierce. US rivals dwarf Mistral, and Chinese open-model developers keep improving. Still, Mensch insists the path forward requires training models in-house, which guarantees customers a steady upgrade path.
By pairing model development with infrastructure ownership, Mistral aims to give Europe a credible, independent contender in the AI race.
