Modelpublic

espressovi/BODHI-gemma-3-math-12b-rlvr

sourceHugging Facegemmaupdated 2mo agoView on Hugging Face

0likes48downloads

Model Card

BODHI

Artifact for the BODHI project. This math model was RL-trained, starting from espressovi/BODHI-gemma-3-12b-distil.

The following results are at T=1.0, 16K tokens and pass@8.

AIME25: 16.57 %