Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Common-sense Reasoning on PIQA, HellaSwag, WinoGrande, ARC-e, ARC-c, SIQA, and BoolQ
Loading...
56.24
Average Accuracy
CCQ-Gated DeltaNet
48.232
50.311
52.39
54.469
May 31, 2026
Average Accuracy
Updated 1d ago
Evaluation Results
Method
Method
Links
Average Accuracy
CCQ-Gated DeltaNet
Scale=1.3B, Training t...
2026.05
56.24
CCQ-GLA
Scale=1.3B, Training t...
2026.05
56.1
GLA-Hedgehog
Scale=1.3B, Training t...
2026.05
55.93
GLA
Scale=1.3B, Training t...
2026.05
55.75
Gated DeltaNet
Scale=1.3B, Training t...
2026.05
55.64
Mamba2
Scale=1.3B, Training t...
2026.05
55.17
Transformer
Scale=1.3B, Training t...
2026.05
54.6
CCQ-Gated DeltaNet
Scale=500M, Training t...
2026.05
49.87
CCQ-GLA
Scale=500M, Training t...
2026.05
49.73
Transformer
Scale=500M, Training t...
2026.05
49.44
Mamba2
Scale=500M, Training t...
2026.05
49.17
Gated DeltaNet
Scale=500M, Training t...
2026.05
48.98
GLA-Hedgehog
Scale=500M, Training t...
2026.05
48.77
GLA
Scale=500M, Training t...
2026.05
48.54
Feedback
Search any
task
Search any
task