Hugging Face's logo

thinletter
/
harrier-0.6b-vq-clients

Feature Extraction
English
embeddings
retrieval
quantization
vector-quantization
webgpu
query-encoder
asymmetric-retrieval
Model card Files
xet
 Community
3
New discussion

The query prompt's K/V in full precision: +0.01–0.04 nDCG@10 at 1.6–1.8 bits per weight for 2 MB of data (measured, not yet in the files)

2
#3 opened 1 day ago by
honza-rosecky

Vector-quantised harrier-0.6b query clients at 1.8–2.1 bits per weight (105–127 MiB) — summary and links
#2 opened 1 day ago by
honza-rosecky