Model comparison

moonshine-base vs whisper-base

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

From the file· summed bytes, KV per layer

Side by side

moonshine-basewhisper-base
Parameters62M73M
Architecturemoonshinewhisper
Layers6
Native context194
Mixture of expertsnono
Quantizations published519
Smallest quantization0.04 GiB0.03 GiB
Q4_K_M0.05 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
Contextmoonshine-basewhisper-baseRatio
4,096
8,192
16,384
32,768
65,536
131,072