Mistral Says Its New AI Model ‘Le Chonk’ Is the Best Open-Weight Offering Outside of China
With the release of its new trillion-parameter model, Mistral is hoping to demonstrate it’s “still in the race” to build frontier-level artificial intelligence.
Mistral, a French artificial intelligence lab, has unveiled its latest model, Le Chonk—a one trillion-parameter model that rivals leading general purpose AI systems. The open-source model, currently in preview, is set to launch in its final form by the end of the month and will be optimized for coding, cyberdefense, and industry-specific tasks.
Mistral's founder and chief scientist, Guillaume Lample, claims Le Chonk is the most capable open-weight model outside of China, as it's trained from scratch rather than through distillation—a technique where smaller models learn from larger ones. Mistral asserts that Le Chonk is "very, very close" to some proprietary models, potentially closing the performance gap with industry giants like OpenAI and Anthropic.
By offering a competitive alternative to closed-weight models from China, Mistral aims to eliminate any remaining reasons businesses might hesitate to choose open-source AI. The French lab has seen significant growth in recent months, raising a $3.3 billion funding round at a $24 billion valuation and experiencing a 20-fold increase in earnings over the past year.
This success comes amid heightened tensions between the US and its allies, as well as debates over access to frontier-grade AI. Mistral's open-weight model could become increasingly attractive to businesses seeking to reduce dependence on proprietary systems and ensure long-term access to AI technology, regardless of location.
Written by urgent.news from Wired's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
