Mistral says it trained ML4 "from scratch" using 3,800 Nvidia Grace Blackwell GPUs in its data centers in Europe and much of its training data was multilingual (Mistral Blog) — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Mistral launched a public preview of Mistral Large 4, informally known as 'Le Chonk,' a 1-trillion-parameter model it claims tops any open model developed in the US or Europe, with full open weights due October 27 (per VentureBeat).
- Mistral says it trained the model 'from scratch' using 3,800 Nvidia Grace Blackwell GPUs housed in its own data centers in Europe, and that much of the training data was multilingual.
Why it matters: Mistral's 1T-parameter model — trained on 3,800 Nvidia Grace Blackwell GPUs in Europe with multilingual data — positions the Paris-based lab as Europe's most ambitious open-weights contender, with the October 27 weight release putting the model in researchers' hands within weeks.
Ask SkimNews
