Science Reasoning on MMLU v1 (test)
79.48AccuracyNEMOTRON-NANO-12B-V2 (RLP)
Evaluation results
| Method | Links | |
|---|---|---|
| NEMOTRON-NANO-12B-V2 (RLP)Pre-training objective=RLP, Post-training pipeline=None, Evaluation protocol=Pass@1 (4 runs)2025.09 | 79.48 | |
| NEMOTRON-NANO-12B-V2 (RLP)Pre-training objective=RLP, Post-training pipeline=None, Evaluation protocol=Greedy2025.09 | 78.76 | |
| NEMOTRON-NANO-12B-V2 (RLP + Post)Pre-training objective=RLP, Post-training pipeline=SFT + RLVR, Evaluation protocol=Greedy2025.09 | 78.17 | |
| NEMOTRON-NANO-12B-V2 (RLP + Post)Pre-training objective=RLP, Post-training pipeline=SFT + RLVR, Evaluation protocol=Pass@1 (4 runs)2025.09 | 77.9 | |
| NEMOTRON-NANO-12B-V2 (Base + Post)Pre-training objective=Next-token prediction, Post-training pipeline=SFT + RLVR, Evaluation protocol=Pass@1 (4 runs)2025.09 | 75.23 | |
| NEMOTRON-NANO-12B-V2 (Base + Post)Pre-training objective=Next-token prediction, Post-training pipeline=SFT + RLVR, Evaluation protocol=Greedy2025.09 | 73.55 | |
| NEMOTRON-NANO-12B-V2 (Base)Pre-training objective=Next-token prediction, Post-training pipeline=None, Evaluation protocol=Greedy2025.09 | 54.12 | |
| NEMOTRON-NANO-12B-V2 (Base)Pre-training objective=Next-token prediction, Post-training pipeline=None, Evaluation protocol=Pass@1 (4 runs)2025.09 | 48.01 |