Is WizardCoder-Python-34B-V1.0 or gemma-4-E4B-it smaller?

gemma-4-E4B-it is smaller at Q4_K_M: 4.64 GiB against 18.83 GiB.

Which uses less memory at long context?

gemma-4-E4B-it — its KV cache at 32K context is 11.9x smaller, because of how its attention layers cache.

Model comparison

WizardCoder-Python-34B-V1.0 vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 20,219,911,424. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 11.9× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	WizardCoder-Python-34B-V1.0	gemma-4-E4B-it
Parameters	33.7B	8.0B
Architecture	llama	gemma4
Layers	48	42
Native context	16,384	131,072
Mixture of experts	no	no
Quantizations published	47	36
Smallest quantization	6.75 GiB	3.30 GiB
Q4_K_M	18.83 GiB	4.64 GiB
Licence	llama2	apache-2.0

KV cache by context

the term that decides long-context viability

Context	WizardCoder-Python-34B-V1.0	gemma-4-E4B-it	Ratio
4,096	0.75 GiB	0.12 GiB	6.10×
8,192	1.50 GiB	0.18 GiB	8.44×
16,384	3.00 GiB	0.29 GiB	10.45×
32,768	6.00 GiB	0.51 GiB	11.86×
65,536	12.00 GiB	0.94 GiB	12.72×
131,072	24.00 GiB	1.82 GiB	13.20×