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

Language Modeling on LM1B (val)

16.57Perplexity

MDM-Prime-v2

11.716844.475977.235109.9941Jun 2, 2025Jul 24, 2025Sep 15, 2025Nov 6, 2025Dec 29, 2025Feb 19, 2026Apr 13, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.03
16.57-
2026.03
20.23-
2025.10
22.32-
2026.03
22.69-
2026.04
22.866.7
2025.10
24.49-
2026.03
26.01-
2026.02
27.04-
2025.10
29.99-
2026.04
3092.2
2026.04
31103.9
2026.04
32115.9
2026.04
32.477.3
2025.10
32.79-
2026.02
32.98-
2026.04
33.697.6
2026.04
33.899.8
2025.10
36.55-
2026.03
38-
2026.03
38.02-
2026.02
39.16-
2026.02
39.98-
2025.10
39.99-
2026.04
40.3-
2026.02
40.7-
2026.03
41.28-
2026.03
43.2-
2026.03
45.36-
2026.02
46.54-
2026.03
51.25-
2025.06
52.14-
2026.03
55.5-
2025.06
60.11-
2026.03
60.23-
2026.03
60.88-
2026.03
60.96-
2026.03
61.14-
2026.03
63.19-
2025.10
63.78-
2026.04
63.8-
2025.06
65.11-
2026.03
66.1-
2026.03
67.01-
2025.06
67.69-
2026.03
68.2-
2025.06
70.79-
2025.06
72.51-
2026.03
73.86-
2025.06
75.75-
2026.04
76.9-
2025.10
77.5-
2025.06
82.01-
2026.04
118.6-
2025.10
118.62-
2026.04
137.9-
2026.04
-96.9