In light of a recent directive from Trump prompting Anthropic to take its AI models off the market, along with increasing demands for independent technology that lessens dependency on the United States, Mistral AI has found itself at the center of attention. However, the French AI company is often misinterpreted, particularly due to its focus on large language models (LLMs).
Evaluating Mistral by its proximity to becoming “the OpenAI of Europe” may lead to unmet expectations. Its chat platform, Vibe (previously known as Le Chat), lacks the brand recognition of ChatGPT, and Claude remains preferred among many Paris-based founders, overshadowing Mistral’s models.
Conversely, many overlook that Mistral, akin to Palantir, employs engineers who collaborate directly with governments and large corporations to integrate AI and customize it for specific needs.
This strategy complements Mistral’s resources—while the company is reportedly aiming to raise about $3.5 billion at a valuation nearing $23.15 billion, which would nearly double its current worth, it still trails behind U.S. rivals. Nonetheless, its revenue is accelerating; in February, Mistral announced an annual recurring revenue exceeding $400 million, up from just $20 million a year prior, and anticipates surpassing $1 billion in ARR this year.
These financial strides have positioned Mistral well in discussions at forums like Davos and in settings where tech leaders often struggle to convey their messages, such as the French Parliament. CEO Arthur Mensch serves as a public advocate for a distinctive vision of AI, but he has much work to do in clearly articulating his company’s goals.
In a detailed post on LinkedIn, Mensch explained Mistral’s primary activities: deploying its models and agent platforms within the infrastructure of its enterprise clients, as well as assisting them in creating custom models via Forge—a platform that utilizes their proprietary data for training.
However, misunderstandings surrounding Mistral arise from its ambitious vision, emphasized by its name, which refers to a strong wind. Mensch declared, “We are dedicated to ensuring that everyone has access to top-notch AI systems, independent of centralized control from governments or corporations seeking to regulate AI deployment.”
This vision extends beyond enterprise solutions; Mistral intends to make substantial investments in research to remain competitive with foundational AI companies. Mensch also shared thoughts on where he believes the company currently stands in this competitive landscape.
“While we do not yet possess the most advanced language models, we are continually closing that gap. We have an exciting model set to launch this summer with open weights, and we plan to offer early access in July. In less computationally intensive fields such as voice, vision, and document processing, we already have state-of-the-art solutions,” Mensch stated.
The upcoming model has garnered some excitement on social media, where Mensch and Mistral investor Marc Andreessen have engaged with humor and shared memes regarding what will not be called “Le Chaton Fat.” This indicates a significant interest from the global audience, particularly “the rest of the world,” in Mistral’s forthcoming innovations.
Perhaps the most intriguing developments are happening beneath the surface. Earlier this year, Mistral acquired Koyeb, an infrastructure startup, to enhance its initiative to create a “true AI cloud.” The company also unveiled a €4 billion investment strategy (around $4.56 billion) aimed at establishing data centers in France and Sweden, reinforcing its commitment to technological sovereignty.
“We are operating under the belief that AI technology is a commodity that every organization needs access to in a secure and cost-effective manner,” Mensch wrote. For those interested in more insights, continue reading.
Who are Mistral AI’s founders?
Mistral’s trio of founders come from a strong background in AI research, having worked with major tech companies in the U.S. that have a presence in Paris. Prior to leading Mistral, Mensch was affiliated with Google’s DeepMind, while CTO Timothée Lacroix and chief scientist Guillaume Lample both worked at Meta.
Mistral has also brought in as co-founding advisers Charles Gorintin and Jean-Charles Samuelian-Werve, co-founders of health insurance startup Alan. Recently, the company appointed three executives to accelerate its growth: Johan Bergqvist as CFO, Brian Hall as CMO, and Kamal Brar as SVP of Partners & Alliances.
What are Mistral AI’s main models?
Mistral has developed an extensive range of models, including LLMs as well as multimodal, reasoning, audio, and OCR models. Not all models prioritize size; among them is Mistral Small 4 and “Les Ministraux,” a series optimized for edge devices like smartphones. Some models feature open weights, and it has made the code agent Leanstral available as open source.
What partnerships has Mistral AI engaged in?
In 2024, Mistral formed a partnership with Microsoft involving a €15 million investment and a strategic alliance to distribute its AI models through Microsoft’s Azure platform.
In May 2025, Mistral announced plans to participate in establishing an AI Campus in the Paris region via a collaboration with UAE investment firm MGX, NVIDIA, and France’s state-owned investment bank Bpifrance.
In June 2025, Mistral revealed plans to create a European platform dedicated to AI, powered by Nvidia processors, called Mistral Compute, scheduled for launch in 2026. This move was praised as “historic” by French President Emmanuel Macron, who appeared alongside Mensch and Nvidia CEO Jensen Huang at the VivaTech conference after the announcement.
In July 2025, the firm introduced AI for Citizens, a project aimed at helping governments and public institutions use AI effectively to enhance public services.
In September 2025, Mistral and chip manufacturer ASML announced a collaboration to explore integrating AI models within ASML’s products and operational processes.
Additionally, Mistral has formed strategic alliances with prominent organizations, including Accenture, Agence France-Presse, the French army, Luxembourg’s job agency, shipping giant CMA, Germany’s defense tech startup Helsing, IBM, Orange, and Stellantis.
How much funding has Mistral AI acquired so far?
Mistral AI’s funding primarily comes from debt financing, although it has also secured several venture funding rounds, totaling approximately $4 billion.
In June 2023, just a month after its inception, Mistral achieved a historic $113 million seed round led by Lightspeed Venture Partners, claimed to be the largest in Europe at the time, with an initial valuation of $260 million.
Other contributors to 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 finalized a €385 million Series A ($415 million) at a valuation of $2 billion, led by Andreessen Horowitz (a16z) with participation from Lightspeed, BNP Paribas, CMA-CGM, Conviction, Elad Gil, General Catalyst, and Salesforce.
Microsoft’s $16.3 million convertible investment in Mistral in February 2024 was considered an extension of the Series A, indicating no change in valuation.
In June 2024, Mistral raised €600 million (approximately $640 million) through a combination of equity and debt, led by General Catalyst at a $6 billion valuation, with key investors including Cisco, IBM, Nvidia, and Samsung Venture Investment Corporation.
In September 2025, Mistral closed a €1.7 billion Series C round (around $2 billion) led by ASML at a valuation of €11.7 billion (about $13.8 billion), with contributions from existing investors DST Global, a16z, Bpifrance, General Catalyst, Index Ventures, Lightspeed, and Nvidia.
What companies has Mistral AI acquired?
Aside from acquiring Koyeb, Mistral has also purchased Emmi, an Austrian firm focusing on physics-related AI, to enhance support for industrial enterprises in their AI transition.
Is Mistral AI looking to produce its own chips?
While Mistral has yet to create its own chips, Mensch hasn’t dismissed the possibility. “Eventually, we may look into designing our own chips, but currently, we depend on Nvidia, which has been an excellent partner for us as we explore several options,” he commented.
What might an exit for Mistral AI entail?
Mensch stated in January 2025 at the World Economic Forum in Davos that Mistral is “not for sale” and indicated that an IPO is in the works.
This perspective aligns with the substantial funding the startup has gathered: Even a potential acquisition by a prospective buyer such as Apple may not yield adequate returns for its investors, especially considering the associated sovereignty concerns.
This narrative was initially published on February 28, 2025, and will be updated periodically.
