Mistral Says Its AI Model ‘Le Chonk’ Has Leapfrogged China in the Open-Weight Race

Mistral says its trillion-parameter open-weight model Le Chonk is now the world's best for automated coding and cybersecurity, after refining an early version it released a few weeks ago.

• The company says the gains came from training the model on increasingly difficult cybersecurity and coding problems, letting it match or surpass models from Chinese labs such as DeepSeek and Z.ai.
• The finalized version is freely available for anyone to access.
• Mistral's VP of science, Pierre Stock, told reporters the release is "a pivotal moment in the global AI leadership race" and offers frontier capabilities for enterprises and governments facing a rising threat level in cyber.
• Mistral argues the West needs its own advanced open-weight models because Chinese ones could act as threat vectors, and because open weights cannot be switched off remotely the way proprietary models can.
• Rafe Pilling of security company Sophos tells WIRED the technology is dual-use, and that any advanced model, whatever its origin, can go rogue without careful supervision.