Model comparison

whisper-small vs whisper-small.en

At Q4_K_M, whisper-small.en is the smaller download — 171,553,856 bytes against 171,630,656.

From the file· summed bytes, KV per layer

Side by side

whisper-smallwhisper-small.en
Parameters242M242M
Architecturewhisperwhisper
Layers1212
Native context
Mixture of expertsnono
Quantizations published196
Smallest quantization0.08 GiB0.16 GiB
Q4_K_M0.16 GiB0.16 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextwhisper-smallwhisper-small.enRatio
4,096
8,192
16,384
32,768
65,536
131,072