Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

STEM Reasoning on Super GPQA

2.096Speedup Ratio

MTP-D

0.97281.26441.5561.8476Mar 25, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.03
2.096
2026.03
2.026
2026.03
1.994
2026.03
1.98
2026.03
1.951
2026.03
1.884
2026.03
1.865
2026.03
1.733
2026.03
1.676
2026.03
1.648
2026.03
1.571
2026.03
1.496
2026.03
1.28
2026.03
1.05
2026.03
1.016