@berryxia: Folks, 3D asset generation is truly taking off. A team led by Assistant Professor Elliott Wu (former Stanford, Oxford background) at the University of Cambridge just released Articraft — a true agentic coding system. It doesn't just generate static 3D models, it...
Summary
The team of Assistant Professor Elliott Wu from the University of Cambridge released Articraft — an agentic coding system that can automatically generate interactive 3D assets with parts, joints, and realistic motion, and open-sourced the Articraft-10K dataset, significantly lowering the asset threshold for robot training and physical AI.
View Cached Full Text
Cached at: 05/16/26, 09:17 AM
Brothers, 3D asset generation is about to take off completely.
Elliott Wu, Assistant Professor at Cambridge (formerly Stanford, Oxford), and his team just released Articraft — a truly agentic coding system.
It doesn’t just generate static 3D models. Instead, it lets an AI agent write its own code, execute it, receive physics validation feedback, iterate, and ultimately output interactive 3D assets with parts, joints, and realistic motion — ready to plug directly into robot simulations and physical AI scenarios.
Previously, creating a high-quality articulated asset took days or even weeks. Now the agent handles everything automatically, and the whole process is ridiculously efficient.
They also open-sourced the Articraft-10K dataset: over 10,000 articulated 3D objects across 250 categories, all interactive and simulation-ready. This drastically lowers the asset barrier for large-scale robot training and physical world AI.
The “data hunger” problem in physical AI and robot simulation has been largely addressed by this agentic pipeline.
Project page: https://articraft3d.github.io
Full code and dataset are open-sourced on GitHub as well.
If you work on robotics, simulation, physical AI, or 3D content generation, this open-source release is worth trying immediately.
PS: Not tested yet — review coming later!
Elliott / Shangzhe Wu (@elliottszwu): Check out Ariticraft 🦾 - a highly efficient agentic system that generates articulated 3D assets fully automatically at scale!
🚀
Similar Articles
@seclink: 5. Open-Source Acceleration of Robot World Models - NVIDIA Cosmos 3 + Isaac GR00T: Physical AI Foundation Models - AGIBOT Genie Sim 3.0: The First Fully Open-Source Robot Simulation Platform (Complete Open Source of Code, Data, and Assets) - VLA (Vision-…
Robot world models and simulation platforms are experiencing open-source acceleration: NVIDIA launched Cosmos 3 and Isaac GR00T physical AI foundation models, AGIBOT released Genie Sim 3.0, a fully open-source simulation platform, VLA models become mainstream for manipulation policies, collectively lowering the entry barrier for the robotics field.
@FakeMaidenMaker: Folks, we've dug up another AI engineering treasure: Hands-On-AI-Engineering. Just open-sourced, it already hit 2.3K stars. One repo stuffed with over 50 real AI projects that you can run directly. The most practical part is it doesn't talk about vague theories; each project is a complete little…
Recommends the freshly open-sourced GitHub repo Hands-On-AI-Engineering with 2.3K stars, containing over 50 runnable AI projects covering RAG, AI agent, OCR, and more. Each project provides full code and instructions, suitable for hands-on learning.
@php_martin: OpenAI Codex is now free. But what really shocked me isn't the free part — it's that local open-source models can deliver AI Agent performance close to the cloud experience. The video demonstrates 4 real-world scenarios: fixing a crashed space game, building a Whac-A-Mole web game in minutes, generating an Apple-style product homepage, and even launching a browser to search, download, and save files on its own.
OpenAI Codex is now free, but even more surprising is that local open-source models can achieve AI Agent performance close to the cloud, demonstrating scenarios like fixing games and developing web games without requiring API keys or internet.
@NFTCPS: Brothers, doing AI without large models is like doing nothing! Today I have to recommend an open-source masterpiece 'Foundations of LLMs' to you. Don't wait, just read it! This book doesn't beat around the bush—it goes deep from the start! From getting started with large language models to architectural evolution, and then it breaks down Prompt engineering, parameter-efficient fine-tuning, model editing, RAG (Retrieval-Augmented Generation) and other hardcore techniques in one go—a one-stop service.
This article promotes the open-source book 'Foundations of LLMs', which systematically explains knowledge about large language models, and introduces the multi-agent development framework Agent-Kernel.
@servasyy_ai: The physics calculations that 3D developers have practiced for years can now be done with just a single sentence. A free tool has been released that generates complex 3D particle effects from English instructions and can immediately output React or three.js code. The barriers of math and physics engines have disappeared, ushering in an era of instantly transforming ideas into reality. Experience
A free tool has been released that can generate complex 3D particle effects from English instructions and directly output React or three.js code, lowering the barrier to 3D physics calculations.