Is llama2-22b-chat-wizard-uncensored or gemma-4-E4B-it smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

gemma-4-E4B-it — its KV cache at 32K context is 64.2x smaller, because of how its attention layers cache.

Model comparison

llama2-22b-chat-wizard-uncensored vs gemma-4-E4B-it

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 64.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	llama2-22b-chat-wizard-uncensored	gemma-4-E4B-it
Parameters	21.8B	8.0B
Architecture	llama	gemma4
Layers	40	42
Native context	2,048	131,072
Mixture of experts	no	no
Quantizations published	2	36
Smallest quantization	7.56 GiB	3.30 GiB
Q4_K_M	—	4.64 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	llama2-22b-chat-wizard-uncensored	gemma-4-E4B-it	Ratio
4,096	4.06 GiB	0.12 GiB	33.02×
8,192	8.13 GiB	0.18 GiB	45.71×
16,384	16.25 GiB	0.29 GiB	56.60×
32,768	32.50 GiB	0.51 GiB	64.25×
65,536	65.00 GiB	0.94 GiB	68.90×
131,072	130.00 GiB	1.82 GiB	71.49×