Games & virtual worlds · PROJECT NOTES
ViZDoom
A configurable first-person game environment exposing Doom-engine observations and controls for visual reinforcement learning and agent evaluation.
What can you do with it?
Create navigation, control or multiplayer scenarios and train agents from screen observations with configurable rewards and game state.
What should you know before using it?
Freedoom assets are supplied; original Doom/Doom II WADs are not bundled. ViZDoom-original code is MIT, while engine components have separate terms. Current wheels require Python 3.10+; macOS 1.3+ wheels target Apple Silicon, and Windows is documented as less tested.
What is verified here?
This entry was reviewed against original project material. We have not independently run the software or reproduced its results. AI-development claims are author-reported; check upstream for current versions, setup requirements, and licenses.
- Resource type
- Environment + benchmark
- Provenance
- Development provenance not established
- Reviewed
- Platforms
- Linux, macOS, Windows