Mistral and NVIDIA Forge Partnership to Accelerate AI in Engineering Workflows
Mistral and NVIDIA have announced a significant collaboration aimed at revolutionizing engineering workflows through the integration of advanced AI models with NVIDIA's accelerated computing and physics modeling tools. This partnership is set to enable faster simulations and facilitate closed-loop, AI-driven design processes across various industries.
Accelerating Engineering Simulations and Design
The initiative will leverage Mistral's AI models, including its latest Mistral Large 4 (Le Chonk), alongside NVIDIA's CUDA-X libraries and NVIDIA Warp, to accelerate engineering simulations. The goal is to create AI-native engineering workflows that can connect simulation, AI model development, and engineering design. This integration is expected to significantly reduce the time engineers spend on complex tasks, allowing for more rapid exploration of design possibilities.
Furthermore, Mistral and NVIDIA will collaborate on integrating Mistral's models with NVIDIA Omniverse libraries. This will enhance GPU-accelerated 3D scene generation and the creation of SimReady assets, streamlining the development of complex digital twins and virtual environments.
Establishing Standards for AI in Engineering
A core component of this collaboration is the development of public benchmarks and open datasets specifically for evaluating physics models in computer-aided engineering (CAE). These resources will provide a common ground for researchers, developers, and industrial teams to compare the accuracy and performance of AI models, and to build and validate automated workflows that convert CAD designs directly into simulations.
Beyond physics AI, the partnership will also focus on creating dedicated evaluations and benchmarks for agentic AI within CAE and electronic design automation (EDA). These benchmarks will assess an AI agent's ability to utilize engineering tools, execute multi-step workflows, and produce accurate results, while also measuring execution time and computational costs. This effort aims to establish consistent and reproducible methods for assessing AI capabilities in engineering, particularly for use cases in aerospace, automotive, and semiconductor industries.
Building on a Strong Foundation
This industrial engineering collaboration extends an existing partnership between Mistral and NVIDIA. The companies have previously worked together on open model development, domain-specific AI, and the necessary infrastructure for production deployment, including the co-development of Mistral NeMo. They are also collaborating within the NVIDIA Nemotron Coalition to advance open-weight frontier models.
Mistral recently previewed Mistral Large 4, its largest multimodal model to date, featuring approximately 1 trillion total parameters and 49 billion active parameters, trained on 3,800 NVIDIA Grace Blackwell GPUs. The company is also expanding its AI infrastructure, aiming for 1 GW of capacity by 2030, further strengthening Europe's position in AI development.
"The next wave of engineering breakthroughs will come from exploring ideas that were once too complex or time-consuming to test," said Tim Costa, vice president and general manager of industrial and computational engineering at NVIDIA. "By combining Mistral’s advanced open models and deep expertise in physics AI with NVIDIA accelerated computing and libraries, we’re helping engineers push the boundaries of design—with shared industry benchmarks to build confidence in what AI can do."
Neil Ashton, VP of Computational Engineering at Mistral, added, "For too long, the pace of engineering has been set by the pace of simulation. With Mistral's open-weight models and physics AI capabilities, combined with NVIDIA’s accelerated computing stack, we’re giving engineers faster simulation cycles, as well as open benchmarks rigorous enough to validate agents on real CAE and EDA work.”
