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

Task Performance Evaluation on PinchBench v1.2.0

90.1Best Score

All INT4

81.228883.531985.83588.1381Apr 24, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
90.188.240.010532.19
2026.04
90.0989.090.011933.21
2026.04
87.6587.080.012734.53
2026.04
85.4684.110.001317.47
2026.04
82.6378.710.001421.8
2026.04
81.5781.260.001619.07