What's Big and What's Small (Let's Find Out)

Explains the differences between big and small pointing out that it is sometimes difficult to tell.

ASIN

B00LDRDPWY

Embedding

CLIP ViT-L/14 · 768d

Distance metric

cosine

Doc fetch

2ms·cache hitGET /v2/namespaces/amazon-products/documents/B00LDRDPWY

How it works

Similar query

21msnearest_to_id → /query

Doc fetch goes through Layer's Aerospike pull-through cache; cache hit served the row without touching turbopuffer. The similar query asks Layer for nearest neighbors of the stored product vector — queries don't go through the doc cache, so no cache header is set.