Model comparison

VibeVoice-1.5B vs orpheus-3b-0.1-ft

At Q4_K_M, orpheus-3b-0.1-ft is the smaller download — 2,092,567,712 bytes against 3,536,105,184.

From the file· summed bytes, KV per layer

Side by side

VibeVoice-1.5Borpheus-3b-0.1-ft
Parameters2.7B3.8B
Architecturevibevoice-asrllama
Layers28
Native context131,072
Mixture of expertsnono
Quantizations published2022
Smallest quantization1.76 GiB0.91 GiB
Q4_K_M3.29 GiB1.95 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextVibeVoice-1.5Borpheus-3b-0.1-ftRatio
4,0960.44 GiB
8,1920.88 GiB
16,3841.75 GiB
32,7683.50 GiB
65,5367.00 GiB
131,07214.00 GiB