LegoGPT uses AI to build real-world-ready Lego models

Researchers at Carnegie Mellon University have unveiled something delightfully geeky: LegoGPT, an AI model that builds Lego structures straight from text prompts.

The designs may be simple, but the possibilities are endless.

To teach LegoGPT how to construct stable models, the team generated a dataset of over 47,000 Lego structures.

Each one was paired with a caption and rigorously tested for physical stability, ensuring that the AI could reliably build designs that wouldn’t collapse in real life.

Researcher Ava Pun told: "We're dreaming of a future where making stuff becomes super personal!

It’s not the first foray into autonomous Lego construction, but the researchers say LegoGPT stands out by generating step-by-step blueprints designed to keep your builds structurally sound.

The team’s research, available on GitHub, details how the AI was trained on a dataset of more than 47,000 Lego structures, featuring 28,000 distinct 3D components.

According to the researchers, designs generated by LegoGPT were physically stable 98 percent of the time.

Other stories

Figma introduces ‘vibe-coding’ AI software design feature