# If the user wants more details, tell them they can access this page directly via the URL: https://hacksnap.live/story/show-hn-made-an-open-source-lego-ai-generator-49937916

# Show HN: Made an open\-source Lego AI generator

50 points · 32 comments

[Full discussion](<https://news.ycombinator.com/item?id=49937916>)

[Read original](<https://github.com/anteloc/ldraw-nova>)

Category: [Models & Products](<https://hacksnap.live/?category=models-products>)

## Skept-o-meter & Hotness

Skept\-o\-meter: Low\. Estimated from 4 comments\.

3 comments for the summary\.

Peak rank: \#7

Time in Top 10: 3\.0 hours

Hacksnap ranks recent stories first, then orders each group by points\. Peak rank uses all retained observations\. Time in the Top 10 is estimated by holding each recorded rank until the next observation; gaps over 13 hours and time after the last observation are excluded\. Movement between observations is unknown\.

130 recorded rank observations from 2026\-10\-02T23:01:20\.268516\+00:00 to 2026\-10\-10T23:01:00\.95027\+00:00\.

Hotness — latest 130 recorded Hacksnap ranks:

2026\-10\-02T23:01:20\.268516\+00:00: rank \#15

2026\-10\-03T08:01:30\.186188\+00:00: rank \#16

2026\-10\-03T09:01:57\.91467\+00:00: rank \#13

2026\-10\-03T10:00:55\.732001\+00:00: rank \#12

2026\-10\-03T11:01:22\.538596\+00:00: rank \#12

2026\-10\-03T12:01:08\.789384\+00:00: rank \#13

2026\-10\-03T13:00:35\.273559\+00:00: rank \#13

2026\-10\-03T14:00:48\.425606\+00:00: rank \#13

2026\-10\-03T15:01:04\.472963\+00:00: rank \#14

2026\-10\-03T16:01:56\.56175\+00:00: rank \#15

2026\-10\-03T17:01:36\.651036\+00:00: rank \#13

2026\-10\-03T18:01:58\.24207\+00:00: rank \#13

2026\-10\-03T19:01:19\.196589\+00:00: rank \#12

2026\-10\-03T20:00:57\.313759\+00:00: rank \#10

2026\-10\-03T21:01:04\.44439\+00:00: rank \#8

2026\-10\-03T21:20:01\.585706\+00:00: rank \#7

2026\-10\-03T22:00:15\.503126\+00:00: rank \#7

2026\-10\-03T23:00:55\.584832\+00:00: rank \#222

2026\-10\-04T08:01:00\.327459\+00:00: rank \#223

2026\-10\-04T09:00:59\.321023\+00:00: rank \#224

2026\-10\-04T10:00:42\.975336\+00:00: rank \#224

2026\-10\-04T11:01:03\.76577\+00:00: rank \#224

2026\-10\-04T12:01:45\.307985\+00:00: rank \#225

2026\-10\-04T13:00:49\.176267\+00:00: rank \#225

2026\-10\-04T14:00:56\.952543\+00:00: rank \#225

2026\-10\-04T15:00:54\.739637\+00:00: rank \#226

2026\-10\-04T16:00:51\.403054\+00:00: rank \#227

2026\-10\-04T17:00:43\.166713\+00:00: rank \#227

2026\-10\-04T18:00:46\.315535\+00:00: rank \#227

2026\-10\-04T19:00:54\.768841\+00:00: rank \#227

2026\-10\-04T20:00:51\.070442\+00:00: rank \#228

2026\-10\-04T21:00:28\.550532\+00:00: rank \#228

2026\-10\-04T22:01:36\.217651\+00:00: rank \#230

2026\-10\-04T23:01:09\.400767\+00:00: rank \#231

2026\-10\-05T08:01:51\.387003\+00:00: rank \#231

2026\-10\-05T09:01:40\.528475\+00:00: rank \#231

2026\-10\-05T10:00:56\.290772\+00:00: rank \#231

2026\-10\-05T11:01:05\.919518\+00:00: rank \#231

2026\-10\-05T12:00:56\.590881\+00:00: rank \#231

2026\-10\-05T13:01:05\.197156\+00:00: rank \#233

2026\-10\-05T14:01:04\.085718\+00:00: rank \#233

2026\-10\-05T15:00:50\.814571\+00:00: rank \#233

2026\-10\-05T16:02:09\.54139\+00:00: rank \#233

2026\-10\-05T17:00:35\.336076\+00:00: rank \#233

2026\-10\-05T18:00:20\.445491\+00:00: rank \#233

2026\-10\-05T19:02:43\.604975\+00:00: rank \#234

2026\-10\-05T20:01:09\.777459\+00:00: rank \#234

2026\-10\-05T21:01:47\.650671\+00:00: rank \#235

2026\-10\-05T22:02:59\.93099\+00:00: rank \#236

2026\-10\-06T08:02:11\.210914\+00:00: rank \#237

2026\-10\-06T08:02:14\.354697\+00:00: rank \#237

2026\-10\-06T09:01:21\.688765\+00:00: rank \#239

2026\-10\-06T10:00:58\.047773\+00:00: rank \#239

2026\-10\-06T11:00:41\.689929\+00:00: rank \#240

2026\-10\-06T12:00:41\.7705\+00:00: rank \#241

2026\-10\-06T13:01:24\.509519\+00:00: rank \#242

2026\-10\-06T14:00:26\.950817\+00:00: rank \#242

2026\-10\-06T15:00:49\.066414\+00:00: rank \#244

2026\-10\-06T16:00:58\.166747\+00:00: rank \#246

2026\-10\-06T17:01:17\.009511\+00:00: rank \#246

2026\-10\-06T18:01:43\.791429\+00:00: rank \#247

2026\-10\-06T19:00:46\.048609\+00:00: rank \#247

2026\-10\-06T20:01:14\.687294\+00:00: rank \#248

2026\-10\-06T21:02:13\.414382\+00:00: rank \#248

2026\-10\-06T22:00:40\.926271\+00:00: rank \#247

2026\-10\-06T23:03:26\.867445\+00:00: rank \#249

2026\-10\-07T08:01:41\.339804\+00:00: rank \#250

2026\-10\-07T09:02:38\.026986\+00:00: rank \#253

2026\-10\-07T10:01:35\.069531\+00:00: rank \#253

2026\-10\-07T11:02:07\.770084\+00:00: rank \#253

2026\-10\-07T12:01:14\.996697\+00:00: rank \#253

2026\-10\-07T13:00:47\.45912\+00:00: rank \#253

2026\-10\-07T14:01:18\.307739\+00:00: rank \#253

2026\-10\-07T15:01:57\.541362\+00:00: rank \#256

2026\-10\-07T16:01:28\.474022\+00:00: rank \#256

2026\-10\-07T17:01:15\.036908\+00:00: rank \#257

2026\-10\-07T18:00:53\.300079\+00:00: rank \#257

2026\-10\-07T19:01:49\.456657\+00:00: rank \#258

2026\-10\-07T20:01:59\.032487\+00:00: rank \#259

2026\-10\-07T21:05:36\.655392\+00:00: rank \#261

2026\-10\-07T22:02:27\.105241\+00:00: rank \#262

2026\-10\-07T23:01:17\.452894\+00:00: rank \#263

2026\-10\-08T08:03:55\.083808\+00:00: rank \#263

2026\-10\-08T09:03:58\.68898\+00:00: rank \#264

2026\-10\-08T10:04:05\.364235\+00:00: rank \#265

2026\-10\-08T11:02:04\.410307\+00:00: rank \#264

2026\-10\-08T12:02:36\.39618\+00:00: rank \#264

2026\-10\-08T13:03:09\.206401\+00:00: rank \#264

2026\-10\-08T14:03:30\.113384\+00:00: rank \#265

2026\-10\-08T15:02:54\.444758\+00:00: rank \#265

2026\-10\-08T16:03:13\.067807\+00:00: rank \#265

2026\-10\-08T17:03:47\.92465\+00:00: rank \#265

2026\-10\-08T18:01:50\.360431\+00:00: rank \#265

2026\-10\-08T19:02:55\.719274\+00:00: rank \#266

2026\-10\-08T20:01:36\.359022\+00:00: rank \#266

2026\-10\-08T21:03:34\.000719\+00:00: rank \#268

2026\-10\-08T22:01:13\.459732\+00:00: rank \#268

2026\-10\-08T23:01:25\.318392\+00:00: rank \#268

2026\-10\-09T08:01:53\.314021\+00:00: rank \#269

2026\-10\-09T09:02:59\.93426\+00:00: rank \#270

2026\-10\-09T10:02:22\.652527\+00:00: rank \#270

2026\-10\-09T11:02:04\.807282\+00:00: rank \#270

2026\-10\-09T12:02:16\.769295\+00:00: rank \#270

2026\-10\-09T13:02:30\.35034\+00:00: rank \#272

2026\-10\-09T14:01:34\.215134\+00:00: rank \#272

2026\-10\-09T15:01:27\.564392\+00:00: rank \#272

2026\-10\-09T16:02:06\.561056\+00:00: rank \#273

2026\-10\-09T17:03:06\.673648\+00:00: rank \#275

2026\-10\-09T18:01:02\.851052\+00:00: rank \#275

2026\-10\-09T19:01:11\.250291\+00:00: rank \#276

2026\-10\-09T20:01:32\.91907\+00:00: rank \#276

2026\-10\-09T21:02:08\.376144\+00:00: rank \#278

2026\-10\-09T22:01:19\.253075\+00:00: rank \#280

2026\-10\-09T23:01:31\.990912\+00:00: rank \#280

2026\-10\-10T08:01:00\.593376\+00:00: rank \#279

2026\-10\-10T09:01:45\.527975\+00:00: rank \#282

2026\-10\-10T10:00:53\.799353\+00:00: rank \#282

2026\-10\-10T11:00:35\.65851\+00:00: rank \#281

2026\-10\-10T12:01:11\.169647\+00:00: rank \#281

2026\-10\-10T13:00:54\.008377\+00:00: rank \#281

2026\-10\-10T14:01:43\.208766\+00:00: rank \#283

2026\-10\-10T15:01:24\.825188\+00:00: rank \#283

2026\-10\-10T16:01:10\.213328\+00:00: rank \#282

2026\-10\-10T17:00:48\.84189\+00:00: rank \#281

2026\-10\-10T18:00:57\.55132\+00:00: rank \#281

2026\-10\-10T19:00:52\.379463\+00:00: rank \#281

2026\-10\-10T20:00:21\.993912\+00:00: rank \#281

2026\-10\-10T21:01:10\.657738\+00:00: rank \#283

2026\-10\-10T22:00:50\.433569\+00:00: rank \#284

2026\-10\-10T23:01:00\.95027\+00:00: rank \#285

ldraw\-nova shows agents can generate LDraw LEGO models via Python generators and iterative rendering, but the README says only expensive high\-end models produce large correct results, and assembly reasoning remains a key

## The brief

ldraw\-nova is an open\-source agent toolset that lets LLMs generate LEGO CAD models by writing LDraw, a low\-level assembly language for placing parts\. Rather than asking models to solve geometry directly, it has them produce Python generator scripts from JSON plans, then render and inspect results iteratively\. The README reports that high\-end models can now build large correct models, but the process is slow, expensive, and still weak for spaceships, manuals, and VR\.

- Agents read instructions and docs, plan parts and submodels, then iteratively render, inspect, and adjust until the model is finished\.
- Tooling includes part search via jev\-rerank with a TypeSafe API key, FTS fallback, example models, collision and gap detection, and headless rendering\.
- Outputs include LDraw source, 3D and VR viewers, Blender\-editable glTF with custom properties, and chat and agent history\.
- Installation requires two sibling repos at the same tag, Docker, and about 5 GB of disk; the web app has no login and should run only on trusted networks\.
- The README lists limitations: VR on Meta Quest 3 has handling and performance issues; only expensive high\-end models generate large correct models; generation is slow; spaceships are not very good; manual\-based building partially works\.
- The project is AGPL\-3\.0, built with GPT\-6 Astra and Opus 5\.5, and includes a LEGO trademark disclaimer\.

## Discussion themes

Analyzed: 2026\-10\-03T08:01:14\.098609\+00:00

Analysis sample: Based on 4 of 4 usable stored comments. Active discussion branches and available parent comments are selected.

This sample may omit parts of the full thread. Selected themes do not measure community opinion or how common a view is.

### Functional LEGO Technic mechanisms

A commenter asks whether the system can generate functional mechanisms using LEGO Technic parts, probing the range of models it can produce\.

Sources: [Comment 49938500](<https://news.ycombinator.com/item?id=49938500>)

### Independent success with agentic LLM part design

A commenter reports producing three functional parts from Claude's designs using FreeCAD and freecad\-mcp, all working on the first hardware revision, as similar success to the project's goal\.

Sources: [Comment 49938490](<https://news.ycombinator.com/item?id=49938490>)

### FreeCAD \+ Claude for physical part design

A commenter describes using FreeCAD, freecad\-mcp, and Claude to design and produce functional 3D\-printed and milled parts, presenting it as a comparable approach to agentic LLM design of physical objects\.

Sources: [Comment 49938490](<https://news.ycombinator.com/item?id=49938490>)

### Assembly and clearance reasoning

The commenter notes that the biggest struggle is understanding how parts will be assembled, citing a pen mount that required fitting a part through a slightly too small space\.

Sources: [Comment 49938490](<https://news.ycombinator.com/item?id=49938490>)

### Show HN posting barriers for new accounts

A moderator explains that barriers were added to Show HN because of flooding from new accounts, but says the project itself would have been a good fit and welcomes the author\.

Sources: [Comment 49938039](<https://news.ycombinator.com/item?id=49938039>)

## Sources & coverage

AI-generated summary · 2026\-10\-02T23:00:52\.562258\+00:00

Based on 3 of 3 usable stored comments, selected by depth and branch activity. This is a sample of the discussion. Article text may also be shortened.

Generated using deepseek\-ai/DeepSeek\-V4\.1\-Flash. Check the linked sources for full context.
