---
license: apache-2.0
datasets:
- microsoft/rStar-Coder
- FreedomIntelligence/medical-o1-reasoning-SFT
- NousResearch/Hermes-3-Dataset
- MegaScience/MegaScience
language:
- hi
- en
- sa
- ru
- ur
- ta
- te
- kn
- mr
metrics:
- accuracy
- brier_score
- character
- code_eval
base_model:
- openai/gpt-oss-20b
pipeline_tag: reinforcement-learning
library_name: flair
---