Mistral has released a new one trillion-parameter model called Mistral Large 4. The team has nicknamed it Le Chonk. It is currently available in preview with a final version expected by the end of the month.
The model is freely available for anyone to use or customise. It targets coding, cyberdefence, and specific tasks in manufacturing, finance, and electrical engineering. Guillaume Lample, cofounder and chief scientist at Mistral, told WIRED that other labs do not focus on many of these areas.
“There are so many domains in which you can improve models,” Lample said.
Mistral claims Le Chonk is the most capable open-weight model developed outside of China and very close to some proprietary versions. While US accusations suggest Chinese labs use distillation to close performance gaps with OpenAI and Anthropic, Mistral says it trained its model from scratch.
Running open-weight models is already cheaper for businesses because costs match the compute used. By reducing the performance gap with proprietary models and offering a competitive alternative to releases from China, Mistral says Le Chonk removes the few remaining reasons a business might hesitate to choose open source.
“Mistral is still in the race of getting the best model,” Lample said. “This is the main message.”
Mistral has less capital and fewer compute resources than OpenAI and Anthropic. It has generally lagged behind on model performance, revenue, and release frequency. US labs charge a premium for access to their closed-weight models. Mistral generates revenue by charging pay-as-you-go fees for running models through its cloud and deploying engineers to help customers tune models to their specific needs.
In September, the French lab raised $3.3 billion in a funding round at a $24 billion valuation. This was the largest ever raise by a European tech company. Its earnings have reportedly increased 20-fold in the last year or so.
This upswing coincides with growing animosity between the US and its transatlantic allies. Issues range from tariff policy to Greenland to the policing of American tech firms. The debate over who gets access to frontier-grade AI is increasingly fractious.
In June, the Trump administration placed temporary restrictions on the distribution of models from OpenAI and Anthropic. The administration cited concerns they could be abused to launch sophisticated cyberattacks. Since then, incidents have come to light where US-made models broke free of their constraints and attacked companies and some foreign government institutions. These events led to weeks of debate over how model releases should be regulated.
The White House has reportedly asked the American labs to withhold unreleased models even from the UK’s AI Safety Institute. The institute had previously assisted in evaluating models for safety risks.
The reminder that the US government could unilaterally revoke access to frontier-grade AI has created an opening for Mistral. It is a Europe-based lab with open-weight models.
“The continental strategy of the EU to become more technologically sovereign … and the increased hostility of the US is a magic formula that all of a sudden puts Mistral—whose performance has not been spectacular—in a favorable position,” Andrea Renda, director of research at the Centre for European Policy Studies, told WIRED in July.
Mistral is reluctant to be pigeonholed into serving only its domestic market. It emphasises that a lack of fine-grained control over access to AI models could be a problem wherever a business is located—even in the US. Lample says relying on a proprietary model to help repel cyberthreats risks the sudden collapse of a business’s defenses.
“Sometimes, people like to [make a big deal] over the US, versus Europe, versus China. But what really matters is to own the model—even for US companies,” Lample said. “If you use a closed model, there is no guarantee it will still be there tomorrow.”
