# GPT-6 Astra cuts stent strain by a third, outpacing industrial optimiser Tosca

A language model agent edited about 1,650 settings in a stent's CAD model, the computer drawing used to build the part. It ran only 15 simulations and improved the design further than a standard industrial tool given the same budget. The work is a preprint and has not been peer reviewed.

- Category: Agents
- Readiness level: II (Benchmarked) on Science Underway's four-step scale, as proposed
- Published: 2026-11-20
- Editor: Stefan Perdal
- URL: https://scienceunderway.com/articles/gpt-6-astra-cuts-stent-strain-by-a-third-outpacing-industrial-optimiser-tosca

## The study

AI agents can redesign engineering parts with more than a thousand CAD parameters and nonlinear simulations (preprint, not peer reviewed). arXiv 2611.26003. https://arxiv.org/abs/2611.26003

## At a glance

- What it shows: An AI language model, working in a loop with design software and a physics simulator, redesigned a dissolvable metal stent. Within 15 simulations it cut strain in the stent's bends by about a third. A widely used industrial tool managed about a tenth.
- What it does not show: It does not show that the new stent works in a body, or that the result repeats. Only one AI model ran on the stent, in one clean run, and everything was tested in simulation.
- Why it matters: Engineers often get only a few dozen slow simulations per design study. A tool that can make large, coordinated changes on that budget could shorten design work on many mechanical parts.

## What the study shows

The authors, from Helmholtz-Zentrum Hereon, Hamburg University of Technology and Saarland University, gave GPT-6 Astra a magnesium stent to improve. A stent is a small mesh tube that holds a narrowed blood vessel open. This one is 39 mm long and dissolves in the body over time. Magnesium is brittle, so strain in the stent's bends, called crowns, decides whether it cracks during deployment.

The agent did not run the simulator itself. It wrote short design files that set values in a CAD model with about 1,650 adjustable values. A fixed pipeline built each design, meshed it and ran a finite-element simulation, a standard way to compute how a part deforms under load. Each simulation took about two hours. The agent then read the full results, wrote its own analysis code and proposed the next designs, up to three at a time.

The goal was to lower a high-end measure of crown strain: the 90th percentile across the 60 crowns. The stent also had to open as wide as the original, shorten no more and use no more metal. In a clean run with 15 simulations, the agent cut the score by 32%. The model calls cost about US$19. About three quarters of the gain came from its first round of three designs.

The authors report that this gain is about ten times larger than the scatter seen when the same design is rebuilt and simulated again. One design from earlier pilot sessions, 25% better on the score, was also checked in a slower, more detailed model of the full deployment. That model included crimping the stent onto a balloon and material damage. There it was 14% less strained than the original, and its worst spot carried 63% less damage at full expansion.

For comparison, the authors ran Tosca, an industrial shape optimiser sold by Dassault Systèmes, with the same 15 simulations. Started from the original design, Tosca improved it by about 10% and then stalled. Started from a simple fix the language model had proposed in one call, Tosca polished it well, reaching 20% below the original overall.

Two smaller tests add context. On Tosca's own bundled example, a plate with a hole, agents built on four different language models beat Tosca in 9 of 12 runs. On a single crown with only four settings, Bayesian optimisation, a standard statistical search method, was a real competitor. The authors conclude that agents lead when the number of design values far exceeds the number of simulations.

## Our reading

In our reading, the interesting part is not that a language model can tweak numbers. It is that the agent chose its own structure. Most of its designs gave whole families of crowns the same shape, then corrected single crowns at the end. That is the step a human engineer normally does before handing a classical optimiser a few parameters. Here the agent did it from the first simulation results.

A shuffle test supports the claim that the agent used the simulation data and not only general knowledge. When per-crown results were secretly swapped between crowns, the agent expected strain where the false numbers put it. In the late rounds, it stopped aiming its changes at the crowns the simulation had actually found most strained. The test is small, two continuations per condition, so it shows a consistent pattern rather than a statistic.

On our readiness scale this sits at level II, benchmarked. The comparisons are careful, with a fixed scoring script and a measured noise level. But the stent is the authors' own test case, not a public benchmark. The code and records are available only on request, and the simulations need commercial Abaqus and Tosca licences.

The rest of the article (what it means, limits, what to watch) is free for members: https://scienceunderway.com/articles/gpt-6-astra-cuts-stent-strain-by-a-third-outpacing-industrial-optimiser-tosca

## How this was made

An AI model read the study and wrote this article in the structure Science Underway uses. An editor reviewed it and approved it for publication. Check the original study before you rely on the details.

---
Source: Science Underway, https://scienceunderway.com/. Please link to the article when you cite it.
