Model comparison

Marco-Nano-Instruct vs Qwen3-8B

At Q4_K_M, Qwen3-8B is the smaller download — 5,027,783,488 bytes against 5,371,989,536. At long context the gap widens: Marco-Nano-Instruct's KV cache at 32K is 1.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Marco-Nano-InstructQwen3-8B
Parameters8.0B8.2B
Architectureqwen3moeqwen3
Layers2836
Native context32,76840,960
Mixture of expertsyes, 232 expertsno
Quantizations published3651
Smallest quantization2.44 GiB2.12 GiB
Q4_K_M5.00 GiB4.68 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextMarco-Nano-InstructQwen3-8BRatio
4,0960.44 GiB0.56 GiB1.29×
8,1920.88 GiB1.13 GiB1.29×
16,3841.75 GiB2.25 GiB1.29×
32,7683.50 GiB4.50 GiB1.29×
65,5367.00 GiB9.00 GiB1.29×
131,07214.00 GiB18.00 GiB1.29×