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

Language Model Evaluation on LM Evaluation Harness (test)

74.22MMLU

E2E

67.959269.584671.2172.8354May 6, 2026
Updated 27d ago

Evaluation Results

MethodLinks
2026.05
74.2256.0166.7288.781.2364.5275.22
2026.05
74.1859.766.3889.0981.464.8375.4
2026.05
74.1757.366.7283.4779.3564.7574.51
2026.05
70.123.8462.2988.6369.9554.6266.85
2026.05
70.0723.8457.8485.1465.6254.8567.56
2026.05
70.0124.0362.0388.3568.7654.7867.17
2026.05
68.5547.0361.4382.480.4754.1177.35
2026.05
68.4447.6961.6982.0280.3954.5376.8
2026.05
68.244.5460.4969.6779.4254.5677.5