Model comparison

Qwen2.5-VL-7B-NSFW-Caption-V3 vs gemma-4-E2B-it

At Q4_K_M, gemma-4-E2B-it is the smaller download — 3,106,738,272 bytes against 4,683,072,384.

From the file· summed bytes, KV per layer

Side by side

Qwen2.5-VL-7B-NSFW-Caption-V3gemma-4-E2B-it
Parameters7.6B5.1B
Architectureqwen2vlgemma4
Layers35
Native context131,072
Mixture of expertsnono
Quantizations published5854
Smallest quantization2.59 GiB2.13 GiB
Q4_K_M4.36 GiB2.89 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen2.5-VL-7B-NSFW-Caption-V3gemma-4-E2B-itRatio
4,0960.05 GiB
8,1920.08 GiB
16,3840.14 GiB
32,7680.25 GiB
65,5360.46 GiB
131,0720.90 GiB