Mistral AI launched Vibe, a rebranded agentic platform for enterprise work and coding, on Thursday while CEO Arthur Mensch told CNBC the company is exploring designing its own chips. The Paris-based firm, valued at nearly €12 billion, is making its most aggressive infrastructure push yet as it tries to close the gap with OpenAI and Anthropic.

“Owning the chips may come, I think it should come at some point, but for now we are relying on Nvidia, which is a great partner to us, and we’re testing a few things here and there,” Mensch told CNBC. Custom chips would let Mistral “lower the cost of deploying tokens to meaningful extents,” he said.

Vibe: Agents for Work and Code

Vibe, formerly Le Chat, combines an AI work assistant and a coding agent in a single product. The work side handles research, document creation, email drafting, and recurring workflow automation across 100+ tool integrations including Slack, GitHub, Jira, and calendar systems. The code side offers full-codebase reasoning, async agents that write and open pull requests overnight, and legacy migration tooling. It supports VS Code, JetBrains, and Zed via ACP, with MCP compatibility for tool connections.

CTO Timothée Lacroix described it as putting “frontier AI to work,” with users able to “set the brief and move on, as Vibe thinks, drafts, and delivers finished work from a single conversation,” according to CNBC.

€4 Billion Infrastructure Bet

Mistral simultaneously announced a new inference-focused data center in Les Ulis, France, with 10 megawatts of computing power, scheduled to open in Q3 2026, according to Rappler. The facility joins two previously announced data centers in France and Sweden as part of a €4 billion investment plan targeting 200 MW of total compute capacity by end of 2027.

“Europe is lagging behind when it comes to the buildout of infrastructure, and so we are investing to close that gap,” Mensch told CNBC. He framed the compute shortage as a macroeconomic issue, saying Europe now views AI as a strategic asset comparable to energy. “You can’t afford to have a commercial deficit of a trillion if you actually want to stay competitive in the race.”

The extra capacity will serve both Mistral’s own customers and other AI labs. “AI labs are in sore need of compute, and we have some of it, and some of them are actually asking us for a lot of compute today,” Mensch said.

Revenue Gap and Vertical Integration

Mistral is targeting €1 billion in 2026 revenue, a fivefold increase from the €200 million it generated the prior year, according to CNBC. That figure still trails the major American labs significantly: OpenAI’s annualized recurring revenue stood at $20 billion in 2025, and Anthropic is on track for $10.9 billion in Q2 2026 revenue.

The simultaneous platform launch, infrastructure buildout, and chip exploration amount to a vertical integration strategy. Mistral is moving from model provider to a company that controls agents, compute, and potentially hardware. Amazon and Google have already designed and deployed custom chips in their data centers, and OpenAI has signaled similar ambitions. Mistral joining that list would make it the first European AI company to pursue full-stack ownership from silicon to agent platform.

The European Competitive Position

Mistral counts ASML among its top enterprise customers and provides AI models to the French military, as Mensch confirmed to Rappler on Thursday. The company pushed back on Pope Leo XIV’s recent encyclical criticizing AI in warfare, with Mensch arguing that “as long as we have adversaries that are threatening, and they are threatening, we do need to have our own capabilities.”

The Vibe launch puts Mistral directly into the enterprise agent market alongside Anthropic’s Claude Code and managed agents, OpenAI’s Codex, and Google’s Antigravity. Whether the €1 billion revenue target and the chip ambitions can close the 10x revenue gap with the American leaders will determine if Europe has a genuine contender in the agent infrastructure race or a well-funded also-ran.