Factoriax
An open-source reinforcement learning research environment inspired by the video game Factorio. Built on GPU parallelization technology, it supports complex long-horizon planning, resource scheduling, and combinatorial optimization tasks, providing a standardized and reproducible testbed for RL algorithm research.
Timeline (last 90 days)
Factoriax is a GPU-parallelized simulation environment inspired by Factorio and built specifically for reinforcement learning research.
Factoriax is an open-source project shared by its developer via Reddit.
Factoriax implements factory grid states and logistics simulation logic in a tensorized manner to accommodate batch parallel computation.
Details about Factoriax such as action/observation interface design, supported algorithm baselines, and performance benchmark data have not been fully disclosed.
All Facts (3)
Factoriax is a GPU-parallelized simulation environment inspired by Factorio and built specifically for reinforcement learning research.
50%UnverifiedFactoriax is an open-source project shared by its developer via Reddit.
50%UnverifiedFactoriax implements factory grid states and logistics simulation logic in a tensorized manner to accommodate batch parallel computation.
50%