Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds
Lumine is an open recipe for generalist agents that completes hours-long 3D open-world missions via end-to-end vision-language modeling with adaptive reasoning and strong cross-game zero-shot generalization.
Published Nov 12, 2025▲ 218 on Hugging FacearXiv ↗
Only vote on papers you've read. Sign in with GitHub to vote.
Lumine delivers a precise, impressive vision-language recipe for long-horizon open-world agents, with striking cross-game zero-shot results, but its single-game evidence, missing baselines, and unreleased training data leave its generalist claims unproven.
Abstract
We introduce Lumine, the first open recipe for developing generalist agents capable of completing hours-long complex missions in real time within challenging 3D open-world environments. Lumine adopts a human-like interaction paradigm that unifies perception, reasoning, and action in an end-to-end manner, powered by a vision-language model. It processes raw pixels at 5 Hz to produce precise 30 Hz keyboard-mouse actions and adaptively invokes reasoning only when necessary. Trained in Genshin Impact, Lumine successfully completes the entire five-hour Mondstadt main storyline on par with human-level efficiency and follows natural language instructions to perform a broad spectrum of tasks in both 3D open-world exploration and 2D GUI manipulation across collection, combat, puzzle-solving, and NPC interaction. In addition to its in-domain performance, Lumine demonstrates strong zero-shot cross-game generalization. Without any fine-tuning, it accomplishes 100-minute missions in Wuthering Waves and the full five-hour first chapter of Honkai: Star Rail. These promising results highlight Lumine's effectiveness across distinct worlds and interaction dynamics, marking a concrete step toward generalist agents in open-ended environments.