{"ok":true,"repo":{"id":584,"full_name":"TianyuCodings/NanoJev","url":"https://github.com/TianyuCodings/NanoJev","description":"A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.","language":"Python","topics":[],"stars":2298,"stars_delta":7,"forks":241,"first_seen":"2026-09-26T18:13:21.566858Z","last_seen":"2026-09-26T21:36:36.218679Z","mentions":0,"platforms":[],"score":0.0,"summary":"NanoJev is a small open-source Python replica of Jev, a 0.6B 'parallel decision' model. Instead of decoding tokens, it takes game states and questions and directly outputs probability distributions over 2–255 candidate choices. The repo includes an end-to-end training pipeline, datasets, and a single checkpoint that plays ViZDoom, Maze, and Snake.","why":"A recent update (Sept 20, 2026) showed one checkpoint beating the original Jev on ViZDoom benchmarks, with side-by-side browser replays comparing NanoJev, Jev, and untuned Qwen — driving attention on trending repos.","audience":"ML researchers and hobbyists interested in token-free decision models, RL-style game-playing agents, and training such systems from scratch.","tags":["llm","reinforcement-learning","python","game-agents","research","open-source"],"summarized_at":"2026-09-26T18:20:41.767661Z","summary_model":"glm-5.3-flash","fetched_at":"2026-09-26T21:37:17.286312Z","meta":{"license":"MIT","archived":false,"homepage":null,"first_stars":2291},"pushed_at":"2026-09-21T09:58:28Z","created_at":"2026-09-17T16:08:30Z","watchers":11,"open_issues":7,"mentions_list":[{"platform":"github","url":"https://github.com/search?q=created%3A%3E2026-09-12&s=stars&o=desc","title":"new repos, most starred","seen_at":"2026-09-26T21:36:36.218913Z"}],"snapshots":[{"captured_at":"2026-09-26T18:20:34.295704Z","stars":2291,"forks":241},{"captured_at":"2026-09-26T18:20:35.655076Z","stars":2291,"forks":241},{"captured_at":"2026-09-26T18:27:28.005287Z","stars":2292,"forks":241},{"captured_at":"2026-09-26T18:29:04.633775Z","stars":2292,"forks":241},{"captured_at":"2026-09-26T18:44:04.338559Z","stars":2292,"forks":241},{"captured_at":"2026-09-26T18:57:18.776304Z","stars":2292,"forks":241},{"captured_at":"2026-09-26T19:12:26.263035Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:15:46.252070Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:19:16.868174Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:21:17.851510Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:27:49.683990Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:31:26.268753Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:34:30.728509Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T19:49:30.785345Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T20:04:36.439430Z","stars":2295,"forks":241},{"captured_at":"2026-09-26T20:19:30.444542Z","stars":2296,"forks":241},{"captured_at":"2026-09-26T20:34:29.993938Z","stars":2298,"forks":241},{"captured_at":"2026-09-26T20:49:30.780738Z","stars":2298,"forks":241},{"captured_at":"2026-09-26T20:52:17.870186Z","stars":2298,"forks":241},{"captured_at":"2026-09-26T21:07:16.997884Z","stars":2298,"forks":241},{"captured_at":"2026-09-26T21:22:17.109466Z","stars":2298,"forks":241},{"captured_at":"2026-09-26T21:37:17.287300Z","stars":2298,"forks":241}]}}