{"slug": "show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone", "title": "Show HN: A 14M-param world model of 2D Minecraft you can play on your phone", "summary": "A developer posting as Maharishiva released tinywmcraft, a 13.9M-parameter world model of the 2D game Craftax-Classic that generates each next frame, reward and episode-end signal from pixels and actions alone, and can be played in a browser or on a phone. The project's leaderboard shows iteration 3 scoring 0.562 ± 0.017, ahead of iteration 2 at 0.463 ± 0.023 and iteration 1 at 0.287 ± 0.019, with training reported to take under 10 minutes for about $0.1 of compute. The release invites coding agents to retrain the model autoresearch-style and beat 0.562, with the fastest in-world diamond run posted before 24 December 2026 (PST) winning a paper copy of Sutton & Barto's Reinforcement Learning.", "body_md": "[GitHub](https://github.com/Maharishiva/tinywmcraft)\n\n# Train a world model in under 10 minutes for $0.1 of compute, and play it!\n\nEvery frame below is generated by a 14M-parameter world model on your GPU, one action at a time. Train your own, [autoresearch style](#autoresearch), and beat ours.\n\nnow playing: [iteration 3](https://github.com/Maharishiva/tinywmcraft/blob/8538aa5/train.py) by [Maharishiva](https://github.com/Maharishiva)\n\n## autoresearch you can play\n\nPoint your coding agent at tinywmcraft: it trains a world model in 10 minutes, scores it, and tries again. Push the score up, then play the model you made: walk around inside it, craft, fight, and see what it gets wrong.\n\nBeat 0.562 and yours is iteration 4: the crown goes next to your GitHub login, and the player on this page switches to your model. Paste this into your coding agent:\n\nThe best so far, by the score: how close imagined play comes to the real game in looks, game logic and memory, and how well the model predicts rewards and deaths, from 0 for a trivial model to 1 for the real game.\n\n1. 1[Maharishiva](https://github.com/Maharishiva) 0.562 ± 0.017[iteration 3](https://github.com/Maharishiva/tinywmcraft/blob/8538aa5/train.py) · train.py: shortcut loss, the heads' gradient into the trunk cut to a tenth in the second half, no hard-window mining, 1/8 game-start windows, two fused kernels, 349 lines\n2. 2[Maharishiva](https://github.com/Maharishiva) 0.463 ± 0.023[iteration 2](https://github.com/Maharishiva/tinywmcraft/blob/80b1c34/train.py) · tile transformer tokenizer trained on clean, half-masked and noised frames, fused Pallas kernels, hard-window mining, half the windows from game starts, symlog two-hot reward head\n3. 3[Maharishiva](https://github.com/Maharishiva) 0.287 ± 0.019[iteration 1](https://github.com/Maharishiva/tinywmcraft/blob/23e6515/train_tiny.py) · train_tiny.py: the whole model in 178 lines, no custom kernels\n\n[the benchmark: score over time, every entry, rules, metrics](benchmark.html)\n\n## \n\n**Get a diamond  inside the world model**, in the player above, in the fewest steps. The fastest run posted before 24 December 2026 (PST) wins a paper copy of Sutton & Barto's *Reinforcement Learning*. Press save clip, then post it in the [discussion](https://github.com/Maharishiva/tinywmcraft/discussions/9) and on X. Rules: [diamond_speedrun.md](https://github.com/Maharishiva/tinywmcraft/blob/main/diamond_speedrun.md). Found a funny glitch? Post it in the [glitch discussion](https://github.com/Maharishiva/tinywmcraft/discussions/10).\n\n### the road to a diamond\n\n## crafting recipe book\n\n| make | needs | where | \n|---|---|---|\n| table | 2 | on an open tile | \n| wood pickaxe | 1 | next to a table; mines stone and coal | \n| wood sword | 1 | next to a table | \n| stone pickaxe | 1 1 | next to a table; mines iron | \n| stone sword | 1 1 | next to a table | \n| furnace | 1 | on an open tile | \n| iron pickaxe | 1 1 1 1 | next to a table and a furnace; mines diamonds | \n| iron sword | 1 1 1 1 | next to a table and a furnace | \n| place stone | 1 | on grass, sand, path, water or lava | \n| plant | 1 | on grass | \n\nWood comes from trees by hand, saplings from grass (1 hit in 10); stone and coal need a wood pickaxe, iron a stone pickaxe, diamonds an iron pickaxe. \"Next to\" is one of the 8 tiles around you. These are the real game's rules: the world model learned them from pixels and does not always keep them. In the real game, the first unlock of each of the 22 achievements is worth +1.\n\n## how it works\n\ntinywmcraft is a world model of [Craftax-Classic](https://github.com/MichaelTMatthews/Craftax), a 2D Minecraft, learned from scratch from pixels and actions alone. Given a history of frames and an action, it generates the next frame, the reward and whether the episode ends; feed its own frames back in and you can play inside it. Every world in the player starts from a first frame the model imagines from noise: nothing is replayed. **13.9M parameters** (5.2M tokenizer, 8.6M world model), trained cold in **under 10 minutes on one A100, about $0.1** of compute.\n\nColab Pro: $9.99 for 100 compute units, and a 40 GB A100 uses about 5.4 an hour (~$0.55/h). A full submission, three seeds with their evals, is under $0.5.\n\n[inside the model: tokenizer, losses, optimizer, kernels, sampling](model.html)", "url": "https://wpnews.pro/news/show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone", "canonical_source": "https://maharishiva.github.io/tinywmcraft/", "published_at": "2026-11-02 18:48:59+00:00", "updated_at": "2026-11-02 19:22:01.170280+00:00", "lang": "en", "topics": ["machine-learning", "artificial-intelligence", "ai-research", "ai-agents", "ai-tools"], "entities": ["Maharishiva", "tinywmcraft", "Craftax-Classic", "GitHub", "Sutton & Barto", "Reinforcement Learning"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone", "markdown": "https://wpnews.pro/news/show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone.md", "text": "https://wpnews.pro/news/show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone.txt", "jsonld": "https://wpnews.pro/news/show-hn-a-14m-param-world-model-of-2d-minecraft-you-can-play-on-your-phone.jsonld"}}