European vendor Mistral AI revealed Tuesday that it raised $2 billion, more than doubling its valuation from $6 billion to $14 billion.
Dutch semiconductor equipment manufacturer ASML Holding NV led the Series C funding round, and existing investors such as Nvidia, General Catalyst and Index Ventures participated.
The funding shows the importance of the AI startup to the European AI market.
A European Champion
Mistral was founded in April 2023 by Arthur Mensch, Guillaume Lample and Timothee Lacroix. Two months after its founding, it raised $113 million; a few months later, it released its first open source language model, Mistral 7B. Since then, the vendor has primarily focused on the enterprise market, although dipped into the consumer market with ChatGPT competitor Le Chat.
“There really isn’t another company in Europe trying to do what they’re trying to do, which is to compete with the large U.S. frontier model providers,” said Nick Patience, an analyst at The Futurum Group. He added that rather than the enterprise in general, Mistral AI has focused specifically on regulated and financial industries while prioritizing model efficiency and smaller language models, which are often deployed on the edge, in the enterprise or on a device.
ASML’s investment is noteworthy, given the vendor is a market leader for manufacturing specialized machines used to produce microchips, according to Patience.
“This is a company that makes high-end machines that build chips; it’s not a logical investment partner,” he said.
It’s more likely that the investment occurred because both ASML and Mistral AI are European champions, according to Patience.
“It represents a statement that Europe is serious about developing competitive AI models,” he said.
The deal also sets up Mistral’s growth in the European AI market, according to Paul Baier, CEO of GAI Insights.
“The fact they’ve done such a great job with both the revenue growth and fundraising is an exception to the rule for Europe,” he said.
A Different Approach
However, it’s not clear if that growth will be enough, Baier added.
“Five years from now, will Mistral be a really important European company or will they be a really important global company?” he said. “That’s solved through sales and distribution.”
Mistral’s approach to generative AI also keeps it from being at the top of the U.S. benchmarks where the company’s models would be compared with models from Anthropic and OpenAI, according to Patience.
“They make performance models, but they’re more interested in building efficient models and in some cases smaller models and then this mix of open and closed source,” Patience said.
Despite its different approach, Mistral has still found a little success in the U.S. market, according to Mark Beccue, an analyst at Omdia, a division of Informa TechTarget.
“Mistral stands on its own,” Beccue said. “They have really good models. They were one of the first to introduce a mixture of expert models, so they compete everywhere.”
He added that despite the growth of sovereign AI, where AI systems and infrastructure could be controlled within a nation’s borders, could lead to an issue because each country might end up prioritizing the models made within its borders.
“Maybe there’s something there, but in my view, Mistral has a worldwide market, and they do very well everywhere,” Beccue continued.
The series funding also shows a contrast between the bullish investment market in the U.S. and the more conservative approach in Europe. While Mistral’s investment is significant, it pales in comparison with the $13 billion Anthropic raised earlier this month in a Series F round.
“The culture of investment is unique in the U.S because it’s just hard to replicate,” Beccue said. “The U.S. culture for those venture capitalists or any kind of investment is to take the risk.”