FrontierThe story, in brief
Mistral AI Releases Mistral Large 4 (Le Chonk): A 1.05T Parameter Multimodal MoE Model
1.05 trillion parameters. Mistral just shipped Le Chonk—and the open weights land October 2026.
The KeyNews take
Why it matters
Mistral Large 4 is a material frontier release: a trillion-scale MoE model with native multimodal, 1M context, and trained on 3,800 Blackwell GPUs in European infrastructure. The API is live now; open weights in 2026 will test whether Europe can compete at scale without US GPU dependency.
The key facts
6 to know1.05 trillion parameters, 49B active (MoE architecture)
Native image input (multimodal)
1 million token context window
Trained on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's European datacenters
API live now; open weights ship end of October 2026
Public preview status (not GA)
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: Mistral AI has released Mistral Large 4, nicknamed Le Chonk, as a public preview. It is a 1.05 trillion parameter Mixture of Experts model with 49 billion active parameters, native image input, and a 1 million token context window, trained on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own…