Compress Vectors by 4x by using 8-bit Rotational Quantizationweaviate.io3 points·bobvanluijt··0 commentsOpen articleSaveView on HN