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

Long-context reasoning on OOLONG

68.4Accuracy

λ-RLM

8.49624.04839.655.152Mar 20, 2026Mar 22, 2026Mar 25, 2026Mar 28, 2026Mar 31, 2026Apr 3, 2026Apr 6, 2026
Updated 10d ago

Evaluation Results

MethodLinks
2026.03
68.4
2026.03
63.8
2026.03
62.5
2026.03
61.7
2026.03
61.2
2026.03
56.5
2026.03
55.8
2026.03
48.3
2026.03
48.2
2026.03
47.1
2026.03
45.7
2026.03
45.6
2026.03
45.3
2026.03
44
2026.03
42.7
42.38
2026.03
40.2
2026.04
39.76
2026.03
38.9
38.08
36.56
2026.04
35.86
2026.04
35.48
2026.03
34.8
2026.04
33.31
2026.04
31.95
2026.03
31.2
2026.04
31.03
2026.04
29.67
2026.03
28.5
2026.03
24.1
2026.03
22.6
2026.03
22.6
2026.03
18.3
2026.03
14.3
2026.03
12.5
2026.03
10.8