Model comparison

Qwen3-Coder-30B-A3B-Instruct vs Fallen-Gemma3-27B-v1

At Q4_K_M, Fallen-Gemma3-27B-v1 is the smaller download — 16,546,688,704 bytes against 18,556,689,568. At long context the gap widens: Fallen-Gemma3-27B-v1'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

Qwen3-Coder-30B-A3B-InstructFallen-Gemma3-27B-v1
Parameters30.5B27.4B
Architectureqwen3moegemma3
Layers4848
Native context262,144
Mixture of expertsyes, 128 expertsno
Quantizations published4625
Smallest quantization7.46 GiB7.86 GiB
Q4_K_M17.28 GiB15.41 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
ContextQwen3-Coder-30B-A3B-InstructFallen-Gemma3-27B-v1Ratio
4,0960.38 GiB0.67 GiB1.80×
8,1920.75 GiB0.91 GiB1.21×
16,3841.50 GiB1.38 GiB1.09×
32,7683.00 GiB2.31 GiB1.30×
65,5366.00 GiB4.19 GiB1.43×
131,07212.00 GiB7.94 GiB1.51×