What Do CAE Simulation Agents Really Need Beyond a Generic Harness?
cs.CE, cs.CL, physics.comp-ph
Submitted: 2026-09-03
Updated: 2026-09-03
Code: https://github.com/openai/codex
License: http://creativecommons.org/licenses/by/4.0/
The gist: Computer-aided engineering (CAE) simulation is among the largest and most demanding areas of engineering, where setting up a solver such as OpenFOAM, FEniCS, or COMSOL takes real expertise.
Terminology
Abstract
Computer-aided engineering (CAE) simulation is among the largest and most demanding areas of engineering, where setting up a solver such as OpenFOAM, FEniCS, or COMSOL takes real expertise. Large language model (LLM) agents promise to turn a natural-language request into a working simulation, and recent CAE agents add simulation-specific machinery: multi-agent decomposition, domain retrieval, and scripted reflection. That machinery suited weak base models; modern harnesses already supply multi-turn reasoning, tool use, and execution feedback. We ask what a CAE simulation agent still needs beyond a generic harness. With information access and repair budget held fixed, a single-agent harness matches or beats multi-agent specialized systems (FoamBench 96.4% vs. 88.2%). Ablations trace this to capabilities the harness already provides: execution-feedback repair lifts FoamBench from 71.8% with no repair round to 96.4%, while scripted reflection adds nothing. The one input that still helps is domain knowledge supplied as solver tutorials, our largest measured gain (80.9% to 96.4%).
Sources
- SimBench: A Framework for Evaluating and Diagnosing LLM-Based Digital-Twin Generation for Multi-Physics Simulation
- OptMetaOpenFOAM: Large Language Model Driven Chain of Thought for Sensitivity Analysis and Parameter Optimization based on CFD
- MetaOpenFOAM: an LLM-based multi-agent framework for CFD
- MetaOpenFOAM 2.0: Large Language Model Driven Chain of Thought for Automating CFD Simulation and Post-Processing
- ALL-FEM: Agentic Large Language models Fine-tuned for Finite Element Methods
- Fine-tuning a Large Language Model for Automating Computational Fluid Dynamics Simulations
- FEABench: Evaluating Language Models on Multiphysics Reasoning Ability
- RExBench: Can coding agents autonomously implement AI research extensions?
- MechAgents: Large language model multi-agent collaborations can solve mechanics problems, generate new data, and integrate knowledge
- FeaGPT: an End-to-End agentic-AI for Finite Element Analysis
- CFDLLMBench: A Benchmark Suite for Evaluating Large Language Models in Computational Fluid Dynamics
- AgenticSciML: Collaborative Multi-Agent Systems for Emergent Discovery in Scientific Machine Learning
- Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgets
- Confucius Code Agent: Scalable Agent Scaffolding for Real-World Codebases
- Integrating Large Language Models for Automated Structural Analysis
- Automating Structural Engineering Workflows with Large Language Model Agents
- VASP Agent: An Agentic Framework for Autonomous First-principles Calculations
- SimuAgent: An LLM-Based Simulink Modeling Assistant Enhanced with Reinforcement Learning
- A Preliminary Assessment of Coding Agents for CFD Workflows
- Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems
Related papers
- Constrained Sensing and Reliable State Estimation with Shallow Recurrent Decoders on a TRIGA Mark II Reactor
- A finite-element-inspired bipartite graph learned simulator for manufacturability assessment in large-deformation sheet forming
- Non-linear mechanical field reconstruction coupling recurrent neural networks with physics-informed graph neural networks
- Neural-Parameterized Cellular Automata for Wildfire Spread
- A mesh-based neural energy method for the simulation of heterogeneous composites
- Evidence-Unit Fairness and the Limits of Query-Adaptive Sparse-Dense Fusion in Financial Document Retrieval