600% memory footprint reduction on Vector Embeddings Quantization (github.com) 4 points by betto 1y ago ↗ HN
1 comment
[ 2.6 ms ] story [ 11.9 ms ] thread