Physics Problem Solving on Interphyre Catapult level 1.0 (test)
54AccuracyGPT-OSS-120B HEXA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-OSS-120B HEXAModel=GPT-OSS-120B, Agent Strategy=HEXA, Evolving Configuration=Off2On Evolving2026.06 | 54 | 16.2 | |
| GPT-OSS-120B REACTModel=GPT-OSS-120B, Agent Strategy=REACT, Evolving Configuration=Off2On Evolving2026.06 | 0 | 25 |