Is glm-4-9b-chat or gemma-4-E4B-it smaller?

gemma-4-E4B-it is smaller at Q4_K_M: 4.64 GiB against 5.82 GiB.

Which uses less memory at long context?

gemma-4-E4B-it — its KV cache at 32K context is 39.5x smaller, because of how its attention layers cache.

Model comparison

glm-4-9b-chat vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 6,250,926,848. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 39.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	glm-4-9b-chat	gemma-4-E4B-it
Parameters	9.4B	8.0B
Architecture	chatglm	gemma4
Layers	40	42
Native context	—	131,072
Mixture of experts	no	no
Quantizations published	58	36
Smallest quantization	2.89 GiB	3.30 GiB
Q4_K_M	5.82 GiB	4.64 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	glm-4-9b-chat	gemma-4-E4B-it	Ratio
4,096	2.50 GiB	0.12 GiB	20.32×
8,192	5.00 GiB	0.18 GiB	28.13×
16,384	10.00 GiB	0.29 GiB	34.83×
32,768	20.00 GiB	0.51 GiB	39.54×
65,536	40.00 GiB	0.94 GiB	42.40×
131,072	80.00 GiB	1.82 GiB	44.00×