Games & virtual worlds · PROJECT NOTES
BALROG
An evaluation harness for language and vision-language agents across existing interactive games, including text adventures and NetHack-based tasks.
What can you do with it?
Run a common evaluation workflow across games and compare long-horizon reasoning under explicit agent and observation settings.
What should you know before using it?
Requires the bundled environments and post-install setup, plus a local model server or paid API access where chosen. The harness is MIT; underlying games keep their own licenses. It is not a newly AI-built game.
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
- Benchmark suite
- Provenance
- Development provenance not established
- Reviewed
- Platforms
- Linux, macOS