Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Physical Reasoning on PIQA
Loading...
40.9
PIQA Normalized Performance
Dual
14.796
21.573
28.35
35.127
Dec 16, 2025
PIQA Normalized Performance
Updated 4d ago
Evaluation Results
Method
Method
Links
PIQA Normalized Performance
Dual
Alpha (α)=63/64, Data...
2025.12
40.9
Autoregressive
Alpha (α)=1, Data repe...
2025.12
39.4
Dual
Alpha (α)=3/4, Data re...
2025.12
36.1
Autoregressive
Alpha (α)=1, Data repe...
2025.12
33.3
Dual
Alpha (α)=1/8, Data re...
2025.12
28.1
Autoregressive
Alpha (α)=1, Data repe...
2025.12
15.8
Feedback
Search any
task
Search any
task