Mistral is collaborating with NVIDIA on AI-native engineering workflows powered by NVIDIA accelerated computing, the company said on 21 October 2026.
The work will combine Mistral's models with NVIDIA's accelerated computing and physics modelling tools to speed up simulation and support AI-driven design. The companies will also integrate Mistral's models with NVIDIA Omniverse libraries for GPU-accelerated 3D scene generation and SimReady asset creation.
A central focus is public benchmarks and open datasets for evaluating physics models, so that researchers and industrial teams can compare accuracy and validate automated CAD-to-simulation workflows. The partners will also develop benchmarks for agentic AI in computer-aided engineering (CAE) and electronic design automation (EDA), testing whether agents can use engineering tools and complete multistep workflows correctly.
Target sectors include aerospace, automotive and semiconductors.
"For too long, the pace of engineering has been set by the pace of simulation," said Neil Ashton, VP of Computational Engineering at Mistral.
The announcement did not give release dates for the benchmarks or datasets.