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

Long-context Reasoning on LongReason 16k prefill length 1.0 (test)

55.67Accuracy

Full

39.560443.742747.92552.1073Jun 25, 2026
Updated 29d ago

Evaluation Results

MethodLinks
2026.06
55.67
2026.06
55.32
2026.06
54.16
2026.06
53.67
2026.06
53.15
2026.06
52.39
2026.06
52.08
2026.06
51.8
2026.06
51.77
2026.06
50.88
2026.06
50.88
2026.06
50.13
2026.06
49.12
2026.06
48.74
2026.06
48.36
2026.06
48.23
2026.06
47.36
2026.06
47.23
2026.06
46.47
2026.06
45.72
2026.06
45.59
2026.06
45.34
2026.06
45.09
2026.06
44.71
2026.06
43.83
2026.06
43.42
2026.06
43.2
2026.06
43.19
2026.06
42.95
2026.06
42.82
2026.06
42.44
2026.06
42.44
2026.06
41.06
2026.06
40.55
2026.06
40.55
2026.06
40.18