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

Truthfulness on TruthfulQA (Accuracy (%))

97.55Truthfulness Accuracy

MetaCrit + Claude-3.5-Sonnet

-3.38222.821549.02575.2285May 17, 2025Jul 8, 2025Aug 29, 2025Oct 20, 2025Dec 11, 2025Feb 1, 2026Mar 26, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2025.07
97.55
2025.07
95.83
2025.07
94.12
2025.07
93.75
2026.03
92.2
2026.03
90.1
2026.03
89.5
2025.07
89.35
2026.03
88.7
2026.03
81.5
2025.07
80.67
2025.07
80.66
2026.03
80.2
2026.03
78.4
2025.07
77.11
2025.07
76.62
2025.07
75.89
2025.07
75.28
2025.07
70.38
2025.05
57
2025.05
55.21
2025.10
55.14
2025.12
54.35
2025.12
54.12
2025.12
54.11
2025.05
54
2025.12
53.99
2025.12
53.99
2025.10
53.81
2025.05
53.8
2025.12
53.72
2025.12
53.52
2025.05
53.5
2025.10
53.37
2025.05
53
2025.05
52.91
2025.12
52.91
2025.12
52.79
2025.12
52.78
2025.10
52.22
2025.05
52
2025.05
52
2025.10
51.82
2025.05
51.64
2025.05
51.5
2025.12
51.46
2025.10
51.21
2025.05
51
2025.10
50.98
2025.10
50.45
2025.10
49.73
2025.10
49.04
2025.05
49
2025.10
46.15
2025.10
44.32
2025.10
43.87
2025.10
43.56
2025.10
43.01
2025.10
42.63
2025.10
42.31
2025.10
42.28
2025.10
41.85
2025.10
41.64
2025.10
41.43
2025.10
40.83
2025.10
40.54
2025.10
40.17
2025.10
39.89
2025.07
24
2025.07
23
2025.07
20.5
2025.07
20.5
2025.07
20.5
2025.07
14
2025.07
14
2025.07
12
2025.07
11.5
2025.07
11
2025.07
7
2025.07
7
2025.07
6.5
2025.07
2
2025.07
2
2025.07
1.5
2025.07
1.5
2025.07
0.5