🚀 Parametric CAD Bench V3 is here. 100 new tasks — significantly more complex than V2 — across Create, Create + Edit & Drawing → CAD. 🏆 Astra — 56.87% | $317.81 🥈 Fable 5.1 — 56.75% | $1,056.04 🥉 Opus 5 — 50.86% | $1,012.21 A few insights: ✏️ All 6 tasks unsolved by every model were Create + Edit. 📐 Gemini 3.8 & Grok 4.7 are relatively close to the leaders on Text → CAD, but fall much further behind on Drawing → CAD — making multimodal engineering understanding a key differentiator. 💰 Astra costs ~3.3× more/token than Kimi K3, yet total benchmark cost was almost identical ($317.81 vs. $314.55) — showing that for harder agentic tasks, model efficiency can matter more than token price. 🤖 Agent framework also matters: Gemini 3.8 Flash scored 36.19% with mini-swe-agent (w/ multimodal bridge) vs. 27.56% with Antigravity. 📊 Leaderboard: cadbench.ai/ 📄 V3 Report: cadbench.ai/news/cad-bench-v… #CADBench #AICAD #ParametricCADBench #EngineeringAI #AIAgents #FreeCAD #Benchmark #ClaudeFable51 #GPT6Astra #Grok47

The United States is facing an annual shortage of approximately 18,000 engineers. At the same time, demand for infrastructure, manufacturing, energy, and advanced hardware continues to grow. We have already seen how AI coding models and agents can dramatically improve the productivity of software engineers. We believe a similar transformation will happen across engineering—from design and simulation to manufacturing and the broader physical world. But engineering is not simply the next version of coding. Engineering AI must reason about geometry, physics, materials, constraints, manufacturability, and safety. The required level of accuracy is much higher, while high-quality engineering data is far more limited. As a result, the gap between today’s AI capabilities and truly reliable engineering automation remains significant. Still, the early results are promising: At gNucleus AI, within narrowly defined engineering workflows where sufficient high-quality data is available, we are already seeing productivity improvements of more than 10×. Tasks that previously required many hours of repetitive CAD or simulation work can increasingly be completed in a fraction of the time—with engineers remaining in control of the final design and decisions. AI will not replace the engineering judgment required to build reliable products and infrastructure. It can, however, reduce the time engineers spend translating intent into CAD models, simulation setups, and repetitive workflows. In a world with too few engineers and too many engineering challenges, giving engineers their time back may be one of AI’s most important contributions. engineeringinc.acec.org/blog… #EngineeringAI #CAD #AIEngineering #FrontierAI #ManufacturingAI #gNucleusAI #PhysicsAI #PhysicalAI 📷