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

Thunderbird

Benchmarks

Task NameDataset NameSOTA ResultTrend
Computer-use Task ExecutionThunderbird
Success Rate57.8
19
Log anomaly detectionThunderbird
F1 Score96.1
13
Anomaly DetectionThunderbird
AUROC94.84
9
Anomaly DetectionThunderbird Log
AUROC94.84
9
Log ParsingThunderbird Loghub 2.0
Global Accuracy (GA)98.5
6
Showing 5 of 5 rows