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

Model Routing on Composite QA Suite (MathQA, LogiQA, MedQA, PIQA, TruthQA, MMLU, GSM8k, GPQA, ASDiv, SoQA)

66.2Overall Accuracy

KMeans

58.50460.50262.564.498Jan 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
66.256.1249.0274.8687.3152.785.1489.2924.860.137.65
2026.01
64.752.7449.0272.685.0741.8981.8389.2929.264.6531.48
2026.01
64.4757.8152.9475.4287.3147.381.3289.2923.657.5837.65
2026.01
63.3757.8152.9471.1979.855082.1788.3927.643.9430.25
2026.01
62.9749.7943.1457.9186.5752.783.785.7127.460.6130.86
2026.01
62.957.8145.172.686.5755.4182.0989.2921.639.3937.04
2026.01
61.939.6641.1855.6585.8233.7886.2575.8925.265.6629.63
2026.01
61.5757.8145.173.4581.3441.8978.2789.292933.8432.72
2026.01
61.3757.3845.168.6485.8248.6580.4889.2926.829.829.01
2026.01
61.3358.6549.0262.9986.5752.784.7289.2925.810.131.48
2026.01
60.559.9247.0675.4285.0741.8978.2789.292910.130.86
2026.01
60.1758.6549.0272.680.644.5977.2583.0428.423.2332.1
2026.01
59.7357.3845.171.7579.8532.4378.6989.2926.221.7229.01
2026.01
59.658.6550.9872.8879.147.376.8389.2923.825.2530.86
2026.01
59.5759.4939.2272.681.3443.2476.9190.1828.614.6530.25
2026.01
59.4759.4945.174.2977.6143.2476.8389.2928.812.1229.63
2026.01
59.1749.3735.2962.9962.6951.3577.0889.2924.254.5535.8
2026.01
58.858.2341.1872.0379.8535.1475.5588.3928.219.1930.25