Model comparison

Mistral-Small-Instruct-2409 vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 13,341,242,112. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 13.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Mistral-Small-Instruct-2409gemma-4-E4B-it
Parameters22.2B8.0B
Architecturellamagemma4
Layers5642
Native context32,768131,072
Mixture of expertsnono
Quantizations published5236
Smallest quantization4.50 GiB3.30 GiB
Q4_K_M12.42 GiB4.64 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextMistral-Small-Instruct-2409gemma-4-E4B-itRatio
4,0960.88 GiB0.12 GiB7.11×
8,1921.75 GiB0.18 GiB9.85×
16,3843.50 GiB0.29 GiB12.19×
32,7687.00 GiB0.51 GiB13.84×
65,53614.00 GiB0.94 GiB14.84×
131,07228.00 GiB1.82 GiB15.40×