Hi there :-) New on HN, first time posting.
Past year, around December, I started experimenting with making ChatGPT and Claude generate source code in LDraw language.
This LDraw is literally an "assembly" language, a low-level programming language that describes how to assemble LEGO pieces together into models, one placement instruction at a time.
When executed by specific tools, like e.g. LDView, LeoCAD, Studio... these instructions become LEGO CAD models, that can be interacted with, modified, etc.
Or, in other words: one LDraw source file in .mpd or .ldr format is equivalent to one LEGO CAD model.
So, the idea I had was: if I manage for maybe ChatGPT or Claude to generate high-quality LDraw source files... then, they would actually be generating high-quality LEGO CAD models, right?
Then, after months of iterations and trying one thing after the other... it worked!!!
Long story short: using GPT-6 Astra and Opus 5.5, I've managed to create a python toolset, instructions, and docs for agents in general. Now, these can be used by them to generate LDraw models.
I've packed it all as a dockerized web app for others to try and experiment, with several providers (and agents) to choose from: OpenAI, Claude and OpenRouter.
I'd really appreciate feedback and comments, let's see where this goes =)
Nice job. I actually read a paper on Arxiv last year about this exact thing - using LLMs to drive model building through LDraw that might be worth reading [1].
They had some interesting experiments like giving the LLM a fixed set of parts as a constraint and asked it to "build" tools out of those available lego.
I think they were using Opus 4.5 so I'm sure with even more powerful models it'll be even better.
The next step after this is to have AI and bots build a real house. Great work!
Up next: a tool that converts a photograph of your dumped bin of old Lego into a catalogue of all the Lego in your collection..
I'm interested in this for robotics. Probably need physics to do that well.
Thanks for sharing - I'm actually working with my kid on something similar but with physics modeling for First Lego League. Cool to see your approach!
Can it create functional mechanisms using LEGO Technic parts?
Any estimates for token costs to generate these models? Curious about costs for Astra, Opus and Jev for some of the examples you have in the video.
This is great
Fun idea. Congrats on building and releasing!
That said, LLMs encroaching into the pure-imagination world of Lego gives me a bit of the icks...
Hi there :-)
New on HN, first time posting, and (I guess) my post on Show HN was rejected...
Anyway, I hope I'm not bothering anyone by posting this outside Show HN, or posting almost the same thing twice.
Past year, around December, I started experimenting with making ChatGPT and Claude generate source code in LDraw language.
This LDraw is literally an "assembly" language, a low-level programming language that describes how to assemble LEGO pieces together into models, one placement instruction at a time.
When executed by specific tools, like e.g. LDView, LeoCAD, Studio... these instructions become LEGO CAD models, that can be interacted with, modified, etc.
Or, in other words: one LDraw source file in .mpd or .ldr format is equivalent to one LEGO CAD model.
So, the idea I had was: if I manage for maybe ChatGPT or Claude to generate high-quality LDraw source files... then, they would actually be generating high-quality LEGO CAD models, right?
Then, after months of iterations and trying one thing after the other... it worked!!!
Long story short: using GPT-6 Astra and Opus 5.5, I've managed to create a python toolset, instructions, and docs for agents in general. Now, these can be used by them to generate LDraw models.
I've packed it all as a dockerized web app for others to try and experiment, with several providers (and agents) to choose from: OpenAI, Claude and OpenRouter.
I'd really appreciate feedback and comments, let's see where this goes =)
Neat! Perhaps I'll finally get around to doing something with the 51515 Lego set I have laying around collecting dust.
> I did this in order to get agentic LLMs capable of designing buildable, physical things!
I've been experimenting with FreeCAD + freecad-mcp + Claude recently and have been similarly surprised how well this works. I've produced three useful parts from Claude's designs: a 3d printed case for an OLED display + some inputs + an ESP32, and a milled aluminium hotend mount adapter for a 3d printer, and a printed pen + pressure sensor mount to use the printer as a plotter. I just provided part/model numbers plus a couple of caliper measurements Claude wasn't able to find online and went through two or three cad revisions before making the parts. All three worked on the first hardware revision.
The biggest thing I've found it seems to struggle with is understanding how the parts will be assembled. The pen mount in particular was tricky to put together, having a part which needed to fit through a slightly too small space to get to its spot. Although it's hard to blame the AI for that when I didn't notice it either I suppose.