Agentic Task Completion on Terminal-Bench All 2
77Pass@1AHE
Evaluation results
| Method | Links | |
|---|---|---|
| AHEBase Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 77 | |
| TF-GRPOBase Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 72.3 | |
| CodexBase Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 71.9 | |
| NexAU0Base Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 69.7 | |
| ACEBase Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 68.9 | |
| terminus-2Base Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 62.9 | |
| opencodeBase Model=GPT-5.4 (high reasoning), Rollouts (k)=22026.04 | 47.2 |