OOLONG
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
OOLONG
68.4Accuracy
37
OOLONG trec_coarse
86.6Score
28
OOLONG
7.1Latency (s)
27
Oolong-Synth
78.41Accuracy
11
Oolong
68Accuracy
10
Oolong real 2025 (test)
15.1Score
9
Oolong real
0.151Score
9
Oolong Real
37.46Score
9
Oolong Synthetic
71.75Score
8
OOLONG (test)
58.1TREC-Q-coarse Accuracy
6
Oolong-Synthetic 199 samples (evaluation set)
89.77Oolong Score
5
OOLONG
64Accuracy
3
OOLONG-REAL 650 samples (175K bucket)
0.249Average Reward
2