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

LLM Evaluation Suite MMLU, GSM8k, HellaSwag, WinoGrande

64.43MMLU Score

FP16

45.29450.26255.2360.198Sep 27, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
64.4378.0173.4270.0971.49-
2025.09
64.177.87369.971.299.54
2025.09
6477.773.369.671.1599.47
2025.09
6477.973.469.971.399.67
2025.09
63.477.77369.770.9599.19
2025.09
62.0675.0671.2767.9669.0996.64
2025.09
62.0372.9370.9566.4668.0995.25
2025.09
61.7670.3671.0769.9368.2895.51
2025.09
60.872.2571.7470.4868.8296.27
2025.09
60.6669.9870.5567.0167.0593.79
2025.09
60.6270.4370.9968.0367.5294.45
2025.09
60.571.4227.0151.0752.573.44
2025.09
60.2668.7671.0567.866.9793.68
2025.09
60.2271.6570.9266.7767.3994.27
2025.09
60.1970.8970.9768.1967.5694.51
2025.09
59.9164.8269.7765.5965.0290.96
2025.09
59.8372.4870.2766.5467.2894.11
2025.09
59.7968.9269.566.8566.2792.69
2025.09
59.5667.7870.0868.0366.3692.83
2025.09
59.4965.6669.0567.3265.3891.46
2025.09
57.7367.4868.9266.6965.2191.21
2025.09
57.6862.3263.8764.8862.1986.99
2025.09
56.8160.867.364.5662.3787.24
2025.09
56.7571.9569.2466.6166.1492.52
2025.09
56.1764.969.5167.1764.4490.14
2025.09
55.9560.867.5764.8862.387.15
2025.09
55.5857.7768.5664.3361.5686.11
2025.09
54.3467.4869.6966.9364.6190.38
2025.09
53.4860.7351.4458.856.1178.49
2025.09
50.5551.3364.6159.8356.5879.15
2025.09
46.0347.5464.2161.5654.8476.71