Model comparison

Qwen2.5-Omni-3B vs Qwen3-VL-8B-Instruct-abliterated-v1

At Q4_K_M, Qwen2.5-Omni-3B is the smaller download — 2,104,931,648 bytes against 5,027,785,792.

From the file· summed bytes, KV per layer

Side by side

Qwen2.5-Omni-3BQwen3-VL-8B-Instruct-abliterated-v1
Parameters5.5B8.8B
Architectureqwen2vlqwen3vl
Layers
Native context
Mixture of expertsnono
Quantizations published2248
Smallest quantization0.89 GiB1.97 GiB
Q4_K_M1.96 GiB4.68 GiB
Licenceotherapache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen2.5-Omni-3BQwen3-VL-8B-Instruct-abliterated-v1Ratio
4,096
8,192
16,384
32,768
65,536
131,072