Is Fara-7B or gemma-4-E2B-it smaller?

gemma-4-E2B-it is smaller at Q4_K_M: 2.89 GiB against 4.36 GiB.

Which uses less memory at long context?

gemma-4-E2B-it — its KV cache at 32K context is 7.1x smaller, because of how its attention layers cache.

Model comparison

Fara-7B vs gemma-4-E2B-it

At Q4_K_M, gemma-4-E2B-it is the smaller download — 3,106,738,272 bytes against 4,683,072,000. At long context the gap widens: gemma-4-E2B-it's KV cache at 32K is 7.1× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Fara-7B	gemma-4-E2B-it
Parameters	8.3B	5.1B
Architecture	qwen2vl	gemma4
Layers	28	35
Native context	128,000	131,072
Mixture of experts	no	no
Quantizations published	39	54
Smallest quantization	2.59 GiB	2.13 GiB
Q4_K_M	4.36 GiB	2.89 GiB
Licence	mit	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Fara-7B	gemma-4-E2B-it	Ratio
4,096	0.22 GiB	0.05 GiB	4.00×
8,192	0.44 GiB	0.08 GiB	5.33×
16,384	0.88 GiB	0.14 GiB	6.40×
32,768	1.75 GiB	0.25 GiB	7.11×
65,536	3.50 GiB	0.46 GiB	7.53×
131,072	7.00 GiB	0.90 GiB	7.76×