LingBot-World 2.0 is an advanced world model that generates infinite, real-time interactive 3D scenes with versatile actions and an agentic harness for autonomous character and environment control.

Stars

1,196

7-day growth

No data

Forks

73

Open issues

4

License

NOASSERTION

Last updated

2026-07-14

Why it is worth attention

It introduces an unbounded interaction horizon with consistent quality, real-time 720p 60fps video, diverse interactive elements (attacking, archery, magic, etc.), and a novel agentic harness that decouples character planning from environmental synthesis.

Who it is for

  • AI researchers in generative world models
  • Game developers seeking real-time interactive environments
  • Robotics and embodied AI simulation engineers
  • Content creators for interactive media and virtual worlds

Use cases

  • Generating infinite, consistent game-like environments with real-time interaction
  • Providing dynamic simulation backgrounds for embodied agent training
  • Enabling real-time interactive video content creation
  • Demonstrating agentic planning and environment co-generation in world models

Strengths

  • Unbounded interaction horizon without degradation in output quality
  • Real-time inference at 60 fps for 720p video via distilled causal-fast model
  • Rich set of text-driven interactive actions and events
  • Integration of an agentic harness (pilot + director agents) for autonomous behavior and scene evolution

Considerations

  • Non-commercial license (CC BY-NC-SA 4.0) restricts commercial use
  • Heavy dependency on Wan2.2 codebase and multi-GPU setup (8 GPUs required for inference)
  • Only the 14B causal-fast model is released; other variants (causal-pretrain, bidirectional, smaller models) are TODO
  • Deployment code not provided; users must rely on third-party solutions like SGLang or flashdreams

README quick start

Installation

Clone the repo:

Description

Infinite Worlds with Versatile Interactions

Related repositories

Similar projects matched by category, topics, and programming language.

NousResearch
Featured
NousResearch GitHub avatar

hermes-agent

Hermes Agent is an open-source, self-improving AI agent with a built-in learning loop that creates skills from experience, runs across platforms, and supports any LLM provider.

AI & Machine Learning
215,644
littledivy
Featured
littledivy GitHub avatar

mimic

mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.

AI & Machine LearningDeveloper Tools
1,057
joeseesun
Featured
joeseesun GitHub avatar

qiaomu-goal-meta-skill

qiaomu-goal-meta-skill transforms vague requests into a copy-ready Codex /goal command with defaults, verification, boundaries, and pause conditions.

AI & Machine Learning
776