APIGen
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
APIGen
0.858NDCG@10
44
APIGen (test)
1.7smECE
21
APIGen (test)
0.02smECE
21
APIGen sampled (test)
88.1Argument F1 (2 calls)
15
APIGen sampled (test)
99.4Tool Selection F1 (2 calls)
15
APIGen (test)
0.02smECE
9
APIGen-MT (test)
90.18Action Recall
4
APIGen (test)
89.6Score (Single)
2