Model comparison

Qwen3.5-0.8B vs MiniCPM-V-4.6

At Q4_K_M, Qwen3.5-0.8B is the smaller download — 527,502,816 bytes against 529,101,504.

From the file· summed bytes, KV per layer

Side by side

Qwen3.5-0.8BMiniCPM-V-4.6
Parameters873M1.3B
Architectureqwen35qwen35
Layers2424
Native context262,144262,144
Mixture of expertsnono
Quantizations published4215
Smallest quantization0.31 GiB0.47 GiB
Q4_K_M0.49 GiB0.49 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3.5-0.8BMiniCPM-V-4.6Ratio
4,0960.05 GiB0.05 GiB1.00×
8,1920.09 GiB0.09 GiB1.00×
16,3840.19 GiB0.19 GiB1.00×
32,7680.38 GiB0.38 GiB1.00×
65,5360.75 GiB0.75 GiB1.00×
131,0721.50 GiB1.50 GiB1.00×
Qwen3.5-0.8B vs MiniCPM-V-4.6 — size, memory and hardware fit — ossmodeldb