Li-Zn-H/AwesomeWorldModels

AwesomeWorldModels is a 'meta-repository' or curated list. From a competitive intelligence perspective, its defensibility is near zero because it contains no proprietary code, datasets, or unique algorithms. Its value lies entirely in the curation labor and its SEO/star-count visibility (270 stars). While 270 stars indicate it has reached a specific audience of researchers and students over its year-long lifespan, the 0.0/hr velocity suggests the project is currently stagnant or has reached a plateau. Frontier labs like OpenAI, Google DeepMind, and Meta (FAIR) represent a high risk not because they will build a competing 'list,' but because their technical reports and foundational papers (e.g., Sora, RT-2, V-JEPA) effectively define the state-of-the-art that this list merely tracks. The 'displacement horizon' is short because a single high-quality survey paper published on arXiv by a known lab would immediately supersede this GitHub repository as the primary reference for the field. There is no technical moat, and the 'switching cost' for a user to move to a different list is zero.