Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on GSM8k (Acc., log C)
Loading...
88.7
Accuracy
BoBN
18.188
36.494
54.8
73.106
Jul 3, 2026
Accuracy
Log Confidence
Updated 18d ago
Evaluation Results
Method
Method
Links
Accuracy
Log Confidence
BoBN
N=16, Backbone=tulu3-8b
2026.07
88.7
-
BoBN
N=4, Backbone=tulu3-8b
2026.07
86.5
-
BoBN
N=1, Backbone=tulu3-8b
2026.07
85.4
0.43
pi_ref
N=16, Backbone=tulu3-8b
2026.07
78.3
-
URIAL
N=16, Backbone=tulu3-8b
2026.07
73.3
-
URIAL
N=1, Backbone=tulu3-8b
2026.07
70.6
0.46
pi_ref
N=1, Backbone=tulu3-8b
2026.07
69.6
0.47
BoBN
N=16, Backbone=zephyr-7b
2026.07
58.7
-
BoBN
N=4, Backbone=zephyr-7b
2026.07
44.5
-
pi_ref
N=16, Backbone=zephyr-7b
2026.07
43.9
-
URIAL
N=16, Backbone=zephyr-7b
2026.07
41.6
-
BoBN
N=1, Backbone=zephyr-7b
2026.07
31.5
0.63
pi_ref
N=1, Backbone=zephyr-7b
2026.07
21.4
0.72
URIAL
N=1, Backbone=zephyr-7b
2026.07
20.9
0.73
Feedback
Search any
task
Search any
task