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

WeChat Mini-App

Benchmarks

Task NameDataset NameSOTA ResultTrend
GUI Agent Task CompletionWeChat Mini-App Synthetic (train val)
VLM Eval Success Rate71.81
4
GUI Agent Task CompletionWeChat Mini-App Real-World (test)
Success Rate (Human)42.28
4
Showing 2 of 2 rows