Is Parable-Granite-4.1-8B-Claude-Fable-5 or Qwen3-4B smaller?

Qwen3-4B is smaller at Q4_K_M: 2.33 GiB against 4.77 GiB.

Which uses less memory at long context?

Qwen3-4B — its KV cache at 32K context is 1.1x smaller, because of how its attention layers cache.

Model comparison

At Q4_K_M, Qwen3-4B is the smaller download — 2,497,280,256 bytes against 5,116,704,896.

From the file· summed bytes, KV per layer

	Parable-Granite-4.1-8B-Claude-Fable-5	Qwen3-4B
Parameters	8.4B	4.0B
Architecture	granite	qwen3
Layers	40	36
Native context	131,072	40,960
Mixture of experts	no	no
Quantizations published	36	29
Smallest quantization	1.85 GiB	1.01 GiB
Q4_K_M	4.77 GiB	2.33 GiB
Licence	apache-2.0	—

the term that decides long-context viability

Context	Parable-Granite-4.1-8B-Claude-Fable-5	Qwen3-4B	Ratio
4,096	0.63 GiB	0.56 GiB	1.11×
8,192	1.25 GiB	1.13 GiB	1.11×
16,384	2.50 GiB	2.25 GiB	1.11×
32,768	5.00 GiB	4.50 GiB	1.11×
65,536	10.00 GiB	9.00 GiB	1.11×
131,072	20.00 GiB	18.00 GiB	1.11×