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

Comprehensive Reasoning and Knowledge Benchmark (GSM8K, MMLU, HumanEval, ARC, GPQA)

95.41GSM8K

Reasoning

25.594843.719961.84579.9701Jan 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
95.4184.453.3389.6357.2595.2599.3279.8453.378.64
2026.01
91.138136.6789.7153.3996.8898.948255.3376.12
2026.01
89.8473.62083.5425.0295.5998.7781.7636.3667.16
2026.01
86.2813081.3227.1396.2798.5981.8529.9468.03
2026.01
83.0282.223.3384.3123.3993.4898.2481.7841.6767.94
2026.01
81.7379.826.6771.3439.8595.2598.5980.829.9467.11
2026.01
69.7578.42084.3148.695.5998.4180.8352.2869.8
2026.01
64.6771.42088.4154.2394.3496.8880.6540.9167.94
2026.01
28.2815.21089.0224.7495.9398.9481.8545.1854.35