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

Multiple Choice Questions on TinyMMLU

86.8Accuracy

PA-GRPO

52.89661.69870.579.302Mar 22, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.03
86.891.889
2026.03
85.291.486
2026.03
84.39288
2026.03
8189.380
2026.03
80.383.385
2026.03
79.984.184
2026.03
76.58681
2026.03
75.181.475
2026.03
73.288.680
2026.03
71.38174
2026.03
68.979.978
2026.03
67.981.975
2026.03
67.479.672
2026.03
67.377.973
2026.03
63.575.568
2026.03
61.176.565
2026.03
59.769.558
2026.03
56.969.157
2026.03
56.683.776
2026.03
55.672.162
2026.03
54.268.957