Model comparison

Qwen3-Coder-30B-A3B-Instruct vs Kimiko-v2-13B

At Q4_K_M, Kimiko-v2-13B is the smaller download — 7,865,956,288 bytes against 18,556,689,568.

From the file· summed bytes, KV per layer

Side by side

Qwen3-Coder-30B-A3B-InstructKimiko-v2-13B
Parameters30.5B13.0B
Architectureqwen3moellama
Layers48
Native context262,144
Mixture of expertsyes, 128 expertsno
Quantizations published4612
Smallest quantization7.46 GiB5.06 GiB
Q4_K_M17.28 GiB7.33 GiB
Licenceapache-2.0creativeml-openrail-m

KV cache by context

the term that decides long-context viability
ContextQwen3-Coder-30B-A3B-InstructKimiko-v2-13BRatio
4,0960.38 GiB
8,1920.75 GiB
16,3841.50 GiB
32,7683.00 GiB
65,5366.00 GiB
131,07212.00 GiB