Terminal Command Execution on Terminal-Bench Core 1.1
34.17AccuracyPivotRL
Evaluation Results
| Method | Links | |
|---|---|---|
| PivotRLBase Model=Nemotron-3-Super, Post-training Stage=After RL Stage2026.03 | 34.17 | |
| Nemotron-3-SuperPost-training Stage=SFT2026.03 | 23.33 |
| Method | Links | |
|---|---|---|
| PivotRLBase Model=Nemotron-3-Super, Post-training Stage=After RL Stage2026.03 | 34.17 | |
| Nemotron-3-SuperPost-training Stage=SFT2026.03 | 23.33 |