Model comparison

merlinite-7b-lab vs Qwen3.6-27B

At Q4_K_M, merlinite-7b-lab is the smaller download — 4,368,484,992 bytes against 16,547,397,888.

From the file· summed bytes, KV per layer

Side by side

merlinite-7b-labQwen3.6-27B
Parameters7.2B27.8B
Architecturellamaqwen35
Layers3264
Native context32,768262,144
Mixture of expertsnono
Quantizations published140
Smallest quantization4.07 GiB8.74 GiB
Q4_K_M4.07 GiB15.41 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextmerlinite-7b-labQwen3.6-27BRatio
4,0960.25 GiB
8,1920.50 GiB
16,3841.00 GiB
32,7682.00 GiB
65,5364.00 GiB
131,0728.00 GiB