Personally I'm waiting for this to get added to http://ann-benchmarks.com/ http://ann-benchmarks.com/ to see how it measures up across datasets. Edit: Looks li

Personally I'm waiting for this to get added to http://ann-benchmarks.com/ http://ann-benchmarks.com/ to see how it measures up across datasets. Edit: Looks like it's already there for some datasets. Interestingly it is primarily strong for the glove-100 dataset. I wonder if there are parts of scann that (accidentally) exploit structure specific to word vectors.