In the wake of the Trump mandate that prompted Anthropic to take its newest AI models offline and increasing demands for indigenous technology to lessen dependence on the U.S., Mistral AI has found itself amidst a storm of interest. However, the French AI sensation is frequently misinterpreted, and its focus on large language models (LLMs) has complicated the narrative. 

Anyone assessing Mistral based on its proximity to becoming ‘the OpenAI of Europe’ will likely be let down. Its chat and agent Vibe, previously known as Le Chat, holds only a fraction of ChatGPT’s brand recognition, and Claude remains more favored than Mistral’s models, even among entrepreneurs at Station F, Paris’ startup hub.

Conversely, casual observers often overlook that the French decacorn is following the Palantir strategy, utilizing forward-deployed engineers who assist governments and major corporations in adopting AI and customizing it for specific applications.

This strategy also aligns better with Mistral’s resources. While there are rumors that the company is aiming to raise approximately $3.5 billion at a valuation of $23.15 billion—nearly double its existing valuation—it still falls significantly short of U.S. frontier laboratories. Nevertheless, its revenue is on the rise; in February, it revealed that its annual recurring revenue had exceeded $400 million, up from $20 million just a year prior, and claimed it was on course to exceed $1 billion in ARR this year.

This growth has allowed Mistral to secure an invitation to high-profile gatherings like Davos, and even in venues where tech CEOs struggle to communicate their messages, such as the French Parliament. Mistral CEO Arthur Mensch has emerged as a public advocate for a distinct vision of AI, but he still has work to do in promoting his own enterprise.

In an extensive LinkedIn post, Mensch elaborated on what the Paris-based company does “for a living”—deploying its models and agent platform on the infrastructure of its enterprise clients, while assisting them in developing custom models through Forge, a platform that enables them to utilize their own data for training.

Nonetheless, the misconceptions and heightened expectations surrounding Mistral do not arise without basis. Named after a wind, the company pursues an ambitious vision. “We exist to ensure that everyone has access to the best AI systems, beyond the centralized control of states or corporations that feel compelled to regulate the final implementation of AI,” Mensch wrote.

This vision implies that Mistral is looking beyond just enterprise solutions. It also intends to continue making significant investments in research to keep pace with foundational AI competitors—and Mensch’s post also discussed where he believes the company currently stands in that context.

“At present, we do not yet have the leading language models, but we are steadily closing that gap. We have an exciting model set to debut this summer—it will be open-weight, and we’ll provide early access to it in July. In areas that are less compute-intensive, such as voice, vision, and document processing, we offer state-of-the-art solutions,” Mensch asserted.

Mistral’s forthcoming model has already stirred some excitement on X, where Mensch and Mistral supporter Marc Andreessen have engaged in humor and promoted memes about what we now know won’t be referred to as “Le Chaton Fat.” This is another indication that the world—particularly “the rest of the world”—is observing whatever Mistral has in store.

The most intriguing developments may be occurring behind closed doors. Earlier this year, Mistral acquired the infrastructure startup Koyeb to advance its plans to create “a true AI cloud.” The company also disclosed a €4 billion investment strategy (approximately $4.56 billion) aimed at establishing data centers in France and Sweden—evoking themes of sovereignty.

“We’re operating on the premise that AI technology is a commoditized resource that every organization needs a secure and affordable supply of,” Mensch wrote. If you’re interested in learning more, continue reading.

Who are Mistral AI’s founders?

Mistral’s three founders share a history in AI research at leading U.S. tech firms that have a presence in Paris. Before assuming the role of CEO at Mistral, Mensch was part of Google’s DeepMind; CTO Timothée Lacroix and Chief Scientist Guillaume Lample are former Meta employees.

Mistral has also awarded the title of co-founding advisers to the co-founders of health insurance startup Alan, Charles Gorintin, and Jean-Charles Samuelian-Werve (who is also a board member). Additionally, it has recently appointed three new executives to bolster its growth: Johan Bergqvist as Chief Financial Officer, Brian Hall as Chief Marketing Officer, and Kamal Brar as SVP, Partners & Alliances.

What are Mistral AI’s main models?

Mistral has created a diverse array of models that range from LLMs to multimodal, reasoning, audio, and OCR models. Not all of its models prioritize size; there’s the appropriately named Mistral Small 4 and “Les Ministraux,” a collection of models designed for edge devices like smartphones. Some are open weights, and it has also made the code agent Leanstral open source.

What partnerships has Mistral AI secured?

In 2024, Mistral finalized an agreement with Microsoft that involved a €15 million investment and a strategic partnership for disseminating the French company’s AI models via Microsoft’s Azure platform.

In May 2025, Mistral announced its participation in the establishment of an AI Campus in the Paris region, as part of a collaborative venture with UAE investment firm MGX, NVIDIA, and France’s state-owned investment bank Bpifrance.

In June 2025, Mistral declared plans to launch a European platform dedicated to AI and powered by Nvidia processors, named Mistral Compute, in 2026. The initiative was acclaimed as “historic” by France’s president, Emmanuel Macron, who shared the stage with Mensch and Nvidia CEO Jensen Huang at the VivaTech conference shortly after the announcement.

In July 2025, Mistral initiated AI for Citizens, a program that the company asserted could “assist States and public institutions in effectively leveraging AI for their citizens by transforming public services.”

In September 2025, Mistral partnered with chip manufacturer ASML “to explore the application of AI models across ASML’s product range, as well as in research, development, and operations.”

Mistral has also established strategic alliances with numerous entities, including Accenture, press agency Agence France-Presse, the French military, Luxembourg’s job agency, shipping giant CMA, German defense tech startup Helsing, IBM, Orange, and Stellantis.

How much funding has Mistral AI raised thus far?

Most of Mistral AI’s funding to date has involved debt financing, but the company has also completed several venture capital rounds, totaling around $4 billion, according to Crunchbase.

In June 2023, just a month following its inception, Mistral AI secured a record $113 million seed round led by Lightspeed Venture Partners. Sources at the time indicated that this seed round, the largest in Europe’s history, valued the startup at $260 million. 

Other participants in that round included Bpifrance, Eric Schmidt, Exor Ventures, First Minute Capital, Headline, JCDecaux Holding, La Famiglia, LocalGlobe, Motier Ventures, Rodolphe Saadé, Sofina, and Xavier Niel.

Six months later, Mistral completed a €385 million Series A ($415 million at that time), with a reported valuation of $2 billion. This round was led by Andreessen Horowitz (a16z) and included participation from Lightspeed, along with BNP Paribas, CMA-CGM, Conviction, Elad Gil, General Catalyst, and Salesforce.

Microsoft’s $16.3 million convertible investment in Mistral—part of a partnership announced in February 2024—was categorized as a Series A extension, suggesting an unchanged valuation.

In June 2024, Mistral secured €600 million (approximately $640 million) through a combination of equity and debt. This long-predicted round was spearheaded by General Catalyst at a $6 billion valuation, with prominent investors such as Cisco, IBM, Nvidia, and Samsung Venture Investment Corporation participating.

In September 2025, Mistral completed a €1.7 billion Series C round (around $2 billion) led by ASML at a €11.7 billion valuation (about $13.8 billion), with contributions from current backers DST Global, a16z, Bpifrance, General Catalyst, Index Ventures, Lightspeed, and Nvidia.

What companies has Mistral AI purchased?

Alongside acquiring Koyeb, an infrastructure startup, Mistral has also acquired Emmi, an Austrian startup specializing in physics AI, aimed at better assisting industrial firms in their AI transformation.

Will Mistral AI produce its own chips?

While Mistral has not yet created its own chips, Mensch hasn’t dismissed the idea. “Owning the chips may eventually happen; I believe it should occur at some point, but for now, we are depending on Nvidia, which is a fantastic partner for us, and we’re exploring a few options here and there,” he stated to CNBC. 

What might a Mistral AI exit entail?

Mistral is “not for sale,” Mensch mentioned in January 2025 at the World Economic Forum in Davos. “Certainly, [an IPO is] the goal.” 

This aligns with the significant amount the startup has raised to date: Even a transaction with a rumored potential buyer like Apple may not yield substantial enough multiples for its investors, not to mention sovereignty concerns related to the acquirer. 

This article was initially published on February 28, 2025, and will be updated regularly.