Model comparison

MARTHA-2B-Qwen3.5-Omni vs Qwen3.5-4B

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

MARTHA-2B-Qwen3.5-OmniQwen3.5-4B
Parameters1.9B4.7B
Architectureqwen35qwen35
Layers32
Native context262,144
Mixture of expertsnono
Quantizations published2457
Smallest quantization0.67 GiB1.42 GiB
Q4_K_M2.52 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextMARTHA-2B-Qwen3.5-OmniQwen3.5-4BRatio
4,0960.13 GiB
8,1920.25 GiB
16,3840.50 GiB
32,7681.00 GiB
65,5362.00 GiB
131,0724.00 GiB