Model comparison

functionary-medium-v3.2 vs Qwen3.6-27B

At Q4_K_M, Qwen3.6-27B is the smaller download — 16,547,397,888 bytes against 42,520,394,592.

From the file· summed bytes, KV per layer

Side by side

functionary-medium-v3.2Qwen3.6-27B
Parameters70.6B27.8B
Architecturellamaqwen35
Layers8064
Native context131,072262,144
Mixture of expertsnono
Quantizations published1840
Smallest quantization15.60 GiB8.74 GiB
Q4_K_M39.60 GiB15.41 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
Contextfunctionary-medium-v3.2Qwen3.6-27BRatio
4,0960.25 GiB
8,1920.50 GiB
16,3841.00 GiB
32,7682.00 GiB
65,5364.00 GiB
131,0728.00 GiB