Mistral and NVIDIA have joined forces to create a new generation of AI-native engineering workflows, powered by Mistral Large 4 and NVIDIA's accelerated computing. This collaboration aims to revolutionize engineering simulation and design processes.
TL;DR
- Mistral and NVIDIA are developing AI-native engineering workflows to accelerate simulation and design processes.
- The partnership leverages Mistral Large 4, a 1 trillion parameter model, and NVIDIA's CUDA-X libraries and Omniverse for 3D scene generation.
- The collaboration will establish public benchmarks and open datasets for evaluating physics models and agentic AI in engineering.
What happened
Mistral and NVIDIA are collaborating to develop AI-native engineering workflows that combine Mistral's Large 4 model, data generation, and physics modeling pipelines with NVIDIA's accelerated computing stack. This initiative aims to accelerate engineering simulation and enable closed-loop workflows that connect simulation, AI model development, and engineering design.
The partnership will integrate Mistral's models with NVIDIA Omniverse libraries for GPU-accelerated 3D scene generation and SimReady asset creation. Additionally, they will develop public benchmarks and open datasets for evaluating physics models, providing a shared foundation for researchers, developers, and industrial teams to compare model accuracy and performance.
Beyond physics AI, Mistral and NVIDIA will work together to develop dedicated evaluations and benchmarks for agentic AI in computer-aided engineering (CAE) and electronic design automation (EDA). These benchmarks will assess agents' ability to use engineering tools, complete multistep workflows, and produce technically correct results, while measuring execution time and computational cost.
Why it matters
This collaboration has the potential to significantly accelerate engineering simulation cycles, enabling engineers to design and build the next generation of products more efficiently. By establishing consistent, reproducible methods for assessing capabilities and tracking progress, Mistral and NVIDIA are helping to define standards for agentic and physics AI evaluations in engineering.
For developers and startups, this partnership opens up new opportunities to specialize AI for their industries, languages, and regions. The shared foundation provided by the NVIDIA Nemotron Coalition and the open-weight models from Mistral will enable enterprises worldwide to build AI around their own expertise and scale it to serve markets worldwide.
However, the success of this collaboration will depend on the ability to integrate these advanced AI capabilities into existing engineering workflows and tools. Additionally, the establishment of public benchmarks and open datasets will be crucial in driving adoption and ensuring the accuracy and performance of these models.
Key facts
- Mistral Large 4, known as Le Chonk, is a multimodal model with approximately 1 trillion total parameters and 49 billion active parameters.
- The model was trained on 3,800 NVIDIA Grace Blackwell GPUs in Mistral-operated data centers in Europe.
- Mistral and NVIDIA are expanding the infrastructure that powers AI training, customization, and inference, with Mistral targeting 1 GW of capacity by 2030.
- The companies previously co-developed Mistral NeMo and are now working together through the NVIDIA Nemotron Coalition to develop a frontier base model.
- Mistral recently introduced the public preview of Mistral Large 4, with weights planned for release at the end of October.
- The collaboration aims to establish public benchmarks and open datasets for evaluating physics models and agentic AI in engineering.
- Mistral and NVIDIA will integrate Mistral's models with NVIDIA Omniverse libraries for GPU-accelerated 3D scene generation and SimReady asset creation.
Context
This partnership between Mistral and NVIDIA builds on their ongoing collaboration in open model development, domain-specific AI, and the infrastructure needed to bring these capabilities into production. The companies have previously co-developed Mistral NeMo and are now working together through the NVIDIA Nemotron Coalition to develop a frontier base model.
The collaboration comes at a time when the demand for AI-driven engineering solutions is growing rapidly. The ability to accelerate simulation cycles and enable closed-loop workflows can significantly enhance the efficiency and effectiveness of engineering design processes.
By establishing public benchmarks and open datasets, Mistral and NVIDIA are not only driving adoption but also ensuring the accuracy and performance of these models. This initiative has the potential to set new standards for agentic and physics AI evaluations in engineering, benefiting developers, startups, and enterprises worldwide.
