Model comparison

ace-step-v1.5-1d-vae-stable-audio-format vs embeddinggemma-300m

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

From the file· summed bytes, KV per layer

Side by side

ace-step-v1.5-1d-vae-stable-audio-formatembeddinggemma-300m
Parameters169M303M
Architectureacestep-vaegemma-embedding
Layers24
Native context2,048
Mixture of expertsnono
Quantizations published110
Smallest quantization0.31 GiB0.26 GiB
Q4_K_M
Licencemit

KV cache by context

the term that decides long-context viability
Contextace-step-v1.5-1d-vae-stable-audio-formatembeddinggemma-300mRatio
4,0960.04 GiB
8,1920.05 GiB
16,3840.08 GiB
32,7680.14 GiB
65,5360.27 GiB
131,0720.52 GiB