Mistral AI to Build First Data Centre in France: Report

Live

Hardik Jagda

CEO Proximity

AI is becoming central to modern sports broadcasting

Andrew Bonwick

Vice President of Product Development at Relm Insurance

How insurance and risk models are evolving to support satellite connectivity

Madhav Sheth

CEO of Ai+ Smartphone

Bold attempt to design smartphones specifically for Indian consumers.

Stephen Rose

CEO Render Networks

Why execution in the trenches may be the real key to AI readiness

Mistral AI to Build First Data Centre in France: Report

French AI startup Mistral has unveiled plans to invest “several billion euros” in building its first data center in France, aiming to gain full control over data storage and processing capabilities. Mistral co-founder and CEO Arthur Mensch made the announcement ahead of the AI summit in Paris, scheduled for February 10 and 11, 2025, local media reported.

Make Telecom Talk My Trusted Source

Also Read: Mistral AI Unveils New Models for On-Device AI Computing

France’s Push to Become a Data Center Hub

According to Mensch, Mistral’s goal is to manage the entire AI value chain, from hardware infrastructure to software. The French AI lab introduced its AI assistant, Le Chat, last week. Ahead of the summit, France has positioned itself as a prime destination for data centers, leveraging its low-carbon nuclear energy and a range of readily available development sites to attract major investors.

Le Chat: The Fastest AI Assistant

On Thursday, Mistral released an app called Le Chat that it said can respond to user questions with 1,000 words per second. President Emmanuel Macron expressed his support on X for a post from Mistral AI, exclaiming, “Vive Le Chat!”

President Emmanuel Macron Support for Mistal AI

“Powered by the highest-performing, lowest-latency Mistral models and the fastest inference engines on the planet, Le Chat reasons, reflects, and responds faster than any other chat assistant—up to around 1,000 words per second. We call this feature Flash Answers, and it’s currently available in preview to all users,” Mistral AI said in a blog post on February 6, 2025.

Mistral stated, “Le Chat combines the high-quality pre-trained knowledge of Mistral models with recent information balanced across web search, robust journalism, social media, and multiple other sources to provide nuanced, evidence-based responses.”

Also Read: Qualcomm and Mistral AI Partner to Bring Generative AI to Snapdragon Devices

Additionally, Le Chat has image and document understanding capabilities. Furthermore, Mistral introduced a code interpreter in Le Chat, enabling users to perform sandboxed code execution, conduct scientific analysis, create visualizations, and run simulations. “This feature makes Le Chat a practical tool for validating algorithms and exploring data insights,” the company said.

Le Chat’s image generation is powered by Black Forest Labs’ Flux Ultra, an image generation model.

“Aligned with Mistral AI’s mission of democratizing AI, Le Chat offers the vast majority of its features for free (including access to the latest models, journalism, image generation, document uploads, and more), with upgraded limits for power users starting at USD 14.99 per month in the Pro tier,” the blog post said.

Mistral AI’s Global Expansion and IPO Plans

Mistral AI’s co-founder reportedly stated that China’s artificial intelligence breakthroughs offer inspiration for Europe in the global AI race.

“DeepSeek is a player that is very similar to us,” Mensch told Politico in an interview, even going so far as to call it “China’s Mistral.” “They’re certainly better funded, but it’s very similar because we’re about the same age,” he reportedly said.

Mistral is also preparing for an initial public offering (IPO), Mensch announced earlier in an interview with Bloomberg at the World Economic Forum (WEF).

“We’re not for sale,” Mensch told Bloomberg, adding that the company plans to open a new office in Singapore to strengthen its presence in the booming Asia-Pacific market. It is also expanding operations across Europe and the United States, building on its mission to compete with industry leaders like OpenAI.

Mistral Small 3 AI Model

On January 30, 2025, Mistral introduced a small model called Mistral Small 3, a latency-optimized 24-billion-parameter model released under the Apache 2.0 license. The company claims it is its most efficient and versatile model.

“Designed to handle 80 percent of generative AI tasks at unprecedented speeds, Small 3 can be deployed anywhere, making it a game-changer for accelerating GenAI adoption across industries worldwide,” the company said, noting that it has collaborated with Hugging Face, Ollama, Kaggle, Together AI, and Fireworks AI to make Small 3 available on their platforms.

Also Read: SAP Expands Its Partnership with Mistral AI to Broaden Customer Choice

Global AI Investments

On Thursday, the United Arab Emirates committed to constructing a data center within an AI campus valued at up to USD 50 billion, while Canadian investment fund Brookfield pledged EUR 20 billion for multiple facilities.

Found this useful? You can support us.