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

Apex

Benchmarks

Task NameDataset NameSOTA ResultTrend
Hyperspectral UnmixingAPEX
RMSE (Mean)0.046
15
Mathematical ReasoningAPEX Shortlist
pass@15.8
14
Mathematical ReasoningAPEX 2025
Accuracy16.7
14
Mathematical Problem SolvingApex 2025
Score93.75
13
Mathematical Problem SolvingApex Shortlist
Score94.27
13
Hyperspectral UnmixingApex
Spectral Angle Error (Road)6.16
12
Hyperspectral UnmixingApex
Missed Endmembers Count0
12
Mathematical ReasoningApex Shortlist
pass@828.22
6
Hyperspectral UnmixingApex
Road IoU50.3
5
Game performance evaluationApex
F1 Score64.8
5
MathAPEX shortlist
Score (%)32.2
4
Super-ResolutionAPEX Simulated (test)
B5 NRMSE0.051
4
Mathematical ReasoningAPEX
Mean@164.59
2
Showing 13 of 13 rows