{"slug": "show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms", "title": "Show HN: I built a playable 211M-param world model of the Backrooms", "summary": "A developer released a 211M-parameter transformer world model, called noclip, that generates each next frame of a playable Backrooms game from the previous 2.4 seconds of frames and player inputs, with weights published on Hugging Face. The model was trained from scratch in PyTorch on 440 hours of scripted play recorded across a simulator of 18 Backrooms levels, using flow matching in a pretrained autoencoder's latent space, and was fed its own imperfect frames as context to limit drift. Running it requires an NVIDIA GPU with roughly 4 GB of memory; the developer cited Oasis Minecraft as inspiration.", "body_md": "A 211M-param transformer acts as a game engine. It draws every next frame from the last 2.4 seconds and my inputs.\n\nBuilt a simulator of 18 Backrooms levels in PyTorch to generate the data.\n\nRecorded 440 hours of scripted play in it.\n\nTrained the model from scratch on that, using flow matching in a pretrained autoencoder's latent space.\n\nThe model was trained on its own imperfect frames as context so it doesn't drift right away. It recalls a few older frames by position, so a quick look back usually works.\n\nInspired by Oasis Minecraft.\n\nTo try it you need an NVIDIA GPU with ~4 GB.\n\nWeights: [https://huggingface.co/yelarys/noclip](https://huggingface.co/yelarys/noclip)\n\nVideo with sound: [https://x.com/yelaryss](https://x.com/yelaryss)\n\nComments URL: [https://news.ycombinator.com/item?id=50594070](https://news.ycombinator.com/item?id=50594070)\n\nPoints: 1\n\n# Comments: 0", "url": "https://wpnews.pro/news/show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms", "canonical_source": "https://github.com/Yelar/noclip", "published_at": "2026-11-21 22:00:20+00:00", "updated_at": "2026-11-21 22:12:05.101587+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "generative-ai", "ai-research"], "entities": ["noclip", "PyTorch", "Hugging Face", "Oasis Minecraft", "Backrooms"], "also_reported_by": [], "takeaway": "A 211M-parameter transformer can serve as a game engine by predicting each frame from the last 2.4 seconds of frames plus inputs, trained on 440 hours of scripted play and runnable on an NVIDIA GPU with about 4 GB of memory.", "recommendation": null, "alternates": {"html": "https://wpnews.pro/news/show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms", "markdown": "https://wpnews.pro/news/show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms.md", "text": "https://wpnews.pro/news/show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms.txt", "jsonld": "https://wpnews.pro/news/show-hn-i-built-a-playable-211m-param-world-model-of-the-backrooms.jsonld"}}