French AI company Mistral has completed a Mistral AI model release of open weights for its trillion-parameter system Mistral Large 4 (ML4), also nicknamed “Le Chonk.” The firm says businesses and governments can download and deploy the model themselves, with a clear pitch toward cybersecurity work where control over data and systems matters.
Mistral first introduced the model earlier this month. Making the weights public—the numerical values that drive the system—means other companies and government agencies can run ML4 for their own purposes rather than depending only on a closed vendor’s service.
Why Mistral is stressing open weights
Competition between open-weight developers such as Mistral and closed-model providers such as OpenAI and Anthropic has intensified. Mistral argues that organisations should not have to rely on outside AI firms for advanced capabilities. That case is especially relevant in cybersecurity, where institutions often want tighter control over sensitive systems.
Chief executive Arthur Mensch framed the company’s stance as keeping AI “a tool for empowerment, not dominance.” With this Mistral AI model release, Mistral says it can reach “state-of-the-art performance” while using a “fraction of the compute and capital committed by closed labs.”
In a company statement issued Thursday, Mistral said that amid growing cyber threats, ML4 brings “state-of-the-art sovereign cyber defense capabilities to enterprises and governments globally,” and that institutions “can now download, deploy, and own frontier AI to detect, analyze, and respond to threats across code and IT infrastructure.”
How Mistral describes ML4 performance
Mistral says the model’s performance is comparable to Chinese open-weight rival GLM-5.3, previously viewed as among the strongest open-weight systems for cyber-related work. According to the company, ML4 uses 47 percent fewer tokens per task, which it presents as a more economical option than GLM-5.3.
Mistral also claims ML4 outranks all other open models on agentic coding tasks on Artificial Analysis’ Coding Agent Index, an AI benchmarking site. The open weights from this Mistral AI model release are being offered on Hugging Face and through the company’s API.
What company executives say about control and risk
Pierre Stock, Vice President of Science at Mistral, said releasing the weights gives organisations more control over how they use AI to defend themselves. “Concretely, it means that, for instance, as a business company or as a government, you’re able to have access to frontier intelligence to defend yourselves against cyberattacks. Access that can’t be cut off by definition because the model is open-weighted,” he said.
Stock added that the model could help “identify, coordinate, and respond in real time to major security incidents,” and that open sourcing allows companies to deploy it on-premises in their own ecosystems. He argued that as AI helps security teams find bugs faster, the same capability should be shared widely to strengthen defences, because some malicious actors may try to use similar technology to attack.
Separately, Stock addressed worries about rogue AI agents. He acknowledged risks but warned against alarmism, saying the problem is hard yet “solvable” and that the company does not want to “fearmonger people.”
Aasan alfaaz mein
Mistral ne apne bade AI model ML4 ke weights khol diye hain, jise download karke khud chalaya ja sakta hai. Company ke mutabiq yeh model cyber attacks ka pata lagane aur jawab dene mein madadgar ho sakta hai. Open weights is liye zaroori hain ke access bahar wali company band na kar sake. Jo organisations apne servers par chalana chahein, unke liye yeh option ab maujood hai.
