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

Downstream

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal ReasoningDownstream Overall
BoN@8 Accuracy55.22
12
Downstream Task EvaluationDownstream
Throughput (tokens/s)32.43
4
Showing 2 of 2 rows