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

Performance Prediction on ARC 1.2k (test)

1.14MAE

Metabench

1.00361.92432.8453.7657Oct 9, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
1.140.971
2025.10
1.470.971
2025.10
1.720.938
2025.10
1.750.938
2025.10
1.960.937
2025.10
2.110.939
2025.10
2.180.948
2025.10
2.220.921
2025.10
2.30.905
2025.10
2.610.898
2025.10
4.550.708