Model comparison

Ling-mini-2.0 vs Qwen3.6-27B

At Q4_K_M, Ling-mini-2.0 is the smaller download — 9,911,575,072 bytes against 16,547,397,888. At long context the gap widens: Ling-mini-2.0's KV cache at 32K is 1.6× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Ling-mini-2.0Qwen3.6-27B
Parameters16.3B27.8B
Architecturebailingmoe2qwen35
Layers2064
Native context32,768262,144
Mixture of expertsyes, 256 expertsno
Quantizations published3040
Smallest quantization4.34 GiB8.74 GiB
Q4_K_M9.23 GiB15.41 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextLing-mini-2.0Qwen3.6-27BRatio
4,0960.16 GiB0.25 GiB1.60×
8,1920.31 GiB0.50 GiB1.60×
16,3840.63 GiB1.00 GiB1.60×
32,7681.25 GiB2.00 GiB1.60×
65,5362.50 GiB4.00 GiB1.60×
131,0725.00 GiB8.00 GiB1.60×