Model comparison

gemma-4-12B-it vs MiMo-VL-7B-RL

At Q4_K_M, MiMo-VL-7B-RL is the smaller download — 4,684,339,136 bytes against 7,121,861,440. At long context the gap widens: gemma-4-12B-it's KV cache at 32K is 1.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

gemma-4-12B-itMiMo-VL-7B-RL
Parameters12.0B8.3B
Architecturegemma4qwen2vl
Layers4836
Native context262,144128,000
Mixture of expertsnono
Quantizations published5964
Smallest quantization3.92 GiB1.86 GiB
Q4_K_M6.63 GiB4.36 GiB
Licenceapache-2.0mit

KV cache by context

the term that decides long-context viability
Contextgemma-4-12B-itMiMo-VL-7B-RLRatio
4,0960.72 GiB0.56 GiB1.28×
8,1920.97 GiB1.13 GiB1.16×
16,3841.47 GiB2.25 GiB1.53×
32,7682.47 GiB4.50 GiB1.82×
65,5364.47 GiB9.00 GiB2.01×
131,0728.47 GiB18.00 GiB2.13×
gemma-4-12B-it vs MiMo-VL-7B-RL — size, memory and hardware fit — ossmodeldb