reasoning-gym Evaluation

We store evaluation results of reasoning-gym datasets (including llm outputs) in this repository, mainly those from our paper REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards.

Team

• Joe Sharratt (joesharratt1229)
• Abdulhakeem Adefioye (Adefioye)
• Zafir Stojanovski (zafstojano)
• Rich Jones (Miserlou)
• Oliver Stanley (olliestanley)
• Jean Kaddour (JeanKaddour)
• Andreas Koepf (andreaskoepf)

Contact / Contributing / Sponsoring

• You can reach the eval-team in the #reasoning-gym channel of the GPU-Mode discord server.
• We would be very happy about donations in the form of OpenRouter API keys (or other inference API providers)!

Citation

If you use our library or the evaluation results in your work, please cite our paper:

@misc{stojanovski2025reasoninggymreasoningenvironments,
      title={REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards},
      author={Zafir Stojanovski and Oliver Stanley and Joe Sharratt and Richard Jones and Abdulhakeem Adefioye and Jean Kaddour and Andreas Köpf},
      year={2025},
      eprint={2505.24760},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2505.24760},
}