Is Qwen3.5-9B or Huihui-Qwen3-VL-8B-Instruct-abliterated smaller?

Huihui-Qwen3-VL-8B-Instruct-abliterated is smaller at Q4_K_M: 4.68 GiB against 5.24 GiB.

Which uses less memory at long context?

Qwen3.5-9B — its KV cache at 32K context is 4.5x smaller, because of how its attention layers cache.

Model comparison

Qwen3.5-9B vs Huihui-Qwen3-VL-8B-Instruct-abliterated

At Q4_K_M, Huihui-Qwen3-VL-8B-Instruct-abliterated is the smaller download — 5,027,785,248 bytes against 5,627,044,256. At long context the gap widens: Qwen3.5-9B's KV cache at 32K is 4.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.5-9B	Huihui-Qwen3-VL-8B-Instruct-abliterated
Parameters	9.7B	8.8B
Architecture	qwen35	qwen3vl
Layers	32	36
Native context	262,144	262,144
Mixture of experts	no	no
Quantizations published	34	46
Smallest quantization	2.97 GiB	1.97 GiB
Q4_K_M	5.24 GiB	4.68 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	Qwen3.5-9B	Huihui-Qwen3-VL-8B-Instruct-abliterated	Ratio
4,096	0.13 GiB	0.56 GiB	4.50×
8,192	0.25 GiB	1.13 GiB	4.50×
16,384	0.50 GiB	2.25 GiB	4.50×
32,768	1.00 GiB	4.50 GiB	4.50×
65,536	2.00 GiB	9.00 GiB	4.50×
131,072	4.00 GiB	18.00 GiB	4.50×