Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Common Sense Reasoning on PIQA (dev)
Loading...
83.2
Accuracy
Megatron-NLG
71.7496
74.7223
77.695
80.6677
Dec 13, 2021
Aug 31, 2022
May 20, 2023
Feb 5, 2024
Oct 24, 2024
Jul 12, 2025
Mar 31, 2026
Accuracy
Updated 17d ago
Evaluation Results
Method
Method
Links
Accuracy
Megatron-NLG
Model Size=530B, Evalu...
2021.12
83.2
GPT-3
Model Size=175B, Evalu...
2021.12
82.3
Gopher
Model Size=280B, Evalu...
2021.12
81.8
GLaM
Model Size=64B/64E, Ev...
2021.12
81.8
GLaM
Model Size=64B/64E, Ev...
2021.12
81.4
GPT-3
Model Size=175B, Evalu...
2021.12
81
GPT-3
Model Size=175B, Evalu...
2021.12
80.5
GLaM
Model Size=64B/64E, Ev...
2021.12
80.4
CKT
Model Scale=large
2023.06
76.07
CALM
Model Scale=large
2023.06
75.11
Content-SharpRouter
Discovery Source=AI Di...
2026.03
74.37
Hier-GateNet
Discovery Source=AI Di...
2026.03
74.37
Gated DeltaNet
Discovery Source=Human...
2026.03
74.1
AdaMulti-PathGateNet
Discovery Source=AI Di...
2026.03
74.1
Mamba2
Discovery Source=Human...
2026.03
73.78
DeltaNet
Discovery Source=Human...
2026.03
73.12
PathGate-FusionNet
Discovery Source=AI Di...
2026.03
72.91
FusionGated-FIRNet
Discovery Source=AI Di...
2026.03
72.91
T5
Model Scale=large
2023.06
72.19
Feedback
Search any
task
Search any
task