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

Report-based data analysis on Deep Data Research (test)

39.15Accuracy (Sample Avg, Msg, MIMIC)

GPT-5-2

27.886830.810933.73536.6591Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
39.1562.2165.9739.7770.5655.9437.262.1565.9337.7270.4655.5755.22
2026.06
37.5953.8563.1535.2870.6448.835.4753.5463.7333.5670.4648.3551.2
2026.06
37.249.9360.1637.0759.6348.5235.1249.8559.6534.7859.6948.5148.34
2026.06
36.7558.6876.5536.9972.9265.6334.658.7776.4534.0872.9265.6257.5
2026.06
36.1860.647737.0558.9266.3434.2660.6276.9235.1259.0866.4155.71
2026.06
36.0958.4670.540.5462.1357.5834.0858.1570.9638.5861.8557.9353.9
2026.06
35.7864.0268.2241.573.9565.5633.5663.6968.4539.2773.8566.2557.84
2026.06
34.956.5475.3234.9256.1854.0132.5356.6275.3533.0456.3154.4751.68
2026.06
32.6353.3358.5737.4558.540.6830.9753.2358.5635.1258.1540.9746.51
2026.06
32.1857.7976.6943.7766.4271.5430.6257.8576.342.3966.4671.957.83
2026.06
30.5155.7667.5135.6160.8857.7228.725667.6633.5660.9258.0851.08
2026.06
30.4949.7850.0230.0445.0727.9928.249.5449.2928.244.6228.138.45
2026.06
28.325969.8443.2765.8158.4625.7859.3869.8640.6665.8558.7153.75