Is Qwen3.5-9B or Fara-7B smaller?

Fara-7B is smaller at Q4_K_M: 4.36 GiB against 5.24 GiB.

Which uses less memory at long context?

Qwen3.5-9B — its KV cache at 32K context is 1.7x smaller, because of how its attention layers cache.

Model comparison

Qwen3.5-9B vs Fara-7B

At Q4_K_M, Fara-7B is the smaller download — 4,683,072,000 bytes against 5,627,044,256. At long context the gap widens: Qwen3.5-9B's KV cache at 32K is 1.7× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.5-9B	Fara-7B
Parameters	9.7B	8.3B
Architecture	qwen35	qwen2vl
Layers	32	28
Native context	262,144	128,000
Mixture of experts	no	no
Quantizations published	34	39
Smallest quantization	2.97 GiB	2.59 GiB
Q4_K_M	5.24 GiB	4.36 GiB
Licence	apache-2.0	mit

KV cache by context

the term that decides long-context viability

Context	Qwen3.5-9B	Fara-7B	Ratio
4,096	0.13 GiB	0.22 GiB	1.75×
8,192	0.25 GiB	0.44 GiB	1.75×
16,384	0.50 GiB	0.88 GiB	1.75×
32,768	1.00 GiB	1.75 GiB	1.75×
65,536	2.00 GiB	3.50 GiB	1.75×
131,072	4.00 GiB	7.00 GiB	1.75×