Princeton-AI2-Lab/Web-World-Models
Official Project Page for Web World Models (https://arxiv.org/abs/2512.23676)
This project helps AI developers build persistent, interactive digital worlds for language agents. It allows for the creation of environments with consistent rules and physics, while still enabling large language models to generate dynamic content, narratives, and high-level decisions. The end-user persona is an AI researcher or developer working on language agents or simulated environments.
Use this if you need to create a stable, rule-bound digital world for language agents, where the environment's core logic is consistent, but narratives and high-level interactions are driven by large language models.
Not ideal if you need a fully generative world where every aspect, including fundamental physics and state, is imagined by a model, or if you're building a traditional web application without AI agent interaction.
Stars
87
Forks
14
Language
TypeScript
License
CC-BY-4.0
Category
Last pushed
Jan 30, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/Princeton-AI2-Lab/Web-World-Models"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
screenpipe/screenpipe
screenpipe turns your computer into a personal AI that knows everything you've done. record....
fiatrete/OpenDAN-Personal-AI-OS
OpenDAN is an open source Personal AI OS , which consolidates various AI modules in one place...
e2b-dev/open-computer-use
AI computer use powered by open source LLMs and E2B Desktop Sandbox
ToGe3688/ioBroker.ai-toolbox
The ioBroker AI Toolbox Adapter is a powerful integration that enables users to create and...
dovvnloading/Cortex
Cortex is a private, secure, and highly responsive desktop AI assistant designed for seamless...