Model comparison

mxbai-embed-xsmall-v1 vs all-MiniLM-L6-v2

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

mxbai-embed-xsmall-v1all-MiniLM-L6-v2
Parameters24M23M
Architecturebertbert
Layers66
Native context4,096512
Mixture of expertsnono
Quantizations published428
Smallest quantization0.03 GiB0.02 GiB
Q4_K_M0.02 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextmxbai-embed-xsmall-v1all-MiniLM-L6-v2Ratio
4,0960.04 GiB0.04 GiB1.00×
8,1920.07 GiB0.07 GiB1.00×
16,3840.14 GiB0.14 GiB1.00×
32,7680.28 GiB0.28 GiB1.00×
65,5360.56 GiB0.56 GiB1.00×
131,0721.13 GiB1.13 GiB1.00×