Mistral AI unveils 1-trillion-parameter Mistral Large 4, plans weight release after safety testing
Summary
Mistral AI launches its 1-trillion-parameter multimodal Mistral Large 4, initially restricting access to a public guardrail endpoint before planning to release the model’s weights in three weeks following safety testing.
Key Points
- Mistral AI releases Mistral Large 4, a 1-trillion-parameter multimodal model called Le Chonk that it positions as a European alternative to U.S. closed models and Chinese open models.
- ML4 is initially available only through a public guardrail endpoint, with Mistral planning to release its weights in three weeks after safety testing and coordination with trusted partners and governments.
- Mistral trains ML4 entirely on its own compute using 4,000 Nvidia GPUs and targets cybersecurity, finance, and chip design use cases.