Model comparison

mxbai-embed-xsmall-v1 vs nomic-embed-text-v1

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: mxbai-embed-xsmall-v1's KV cache at 32K is 4.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

mxbai-embed-xsmall-v1nomic-embed-text-v1
Parameters24M137M
Architecturebertnomic-bert
Layers612
Native context4,0968,192
Mixture of expertsnono
Quantizations published414
Smallest quantization0.03 GiB0.05 GiB
Q4_K_M0.08 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextmxbai-embed-xsmall-v1nomic-embed-text-v1Ratio
4,0960.04 GiB0.14 GiB4.00×
8,1920.07 GiB0.28 GiB4.00×
16,3840.14 GiB0.56 GiB4.00×
32,7680.28 GiB1.13 GiB4.00×
65,5360.56 GiB2.25 GiB4.00×
131,0721.13 GiB4.50 GiB4.00×