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

Long context retrieval and reasoning on RULER

94.7Score

Nemotron 3 Ultra

35.981651.225866.4781.7142Jun 12, 2026Jun 15, 2026Jun 18, 2026Jun 21, 2026Jun 24, 2026Jun 27, 2026Jun 30, 2026
Updated 24d ago

Evaluation Results

MethodLinks
2026.06
94.7
2026.06
94.2
2026.06
90.74
2026.06
90.1
2026.06
88.35
2026.06
87.7
2026.06
87.34
2026.06
84.17
2026.06
83.41
2026.06
82.04
2026.06
81.94
2026.06
81.17
2026.06
80.37
2026.06
79.33
2026.06
79.18
2026.06
78.43
2026.06
78.41
2026.06
77.48
2026.06
76.23
2026.06
74.82
2026.06
74.31
2026.06
73.67
2026.06
73.47
2026.06
72.83
2026.06
71.24
2026.06
70.14
2026.06
68.42
2026.06
68.38
2026.06
67.82
2026.06
63.84
2026.06
61.87
2026.06
61.33
2026.06
61.23
2026.06
60.43
2026.06
59.12
2026.06
58.43
2026.06
58.34
2026.06
57.84
2026.06
56.33
2026.06
56.17
2026.06
55.72
2026.06
54.83
2026.06
53.41
2026.06
52.17
2026.06
51.63
2026.06
48.71
2026.06
45.17
2026.06
41.32
2026.06
38.24