Model comparison

Olmo-3.1-32B-Instruct vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 19,482,034,976. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 5.6× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Olmo-3.1-32B-Instructgemma-4-E4B-it
Parameters32.2B8.0B
Architectureolmo2gemma4
Layers6442
Native context65,536131,072
Mixture of expertsnono
Quantizations published4936
Smallest quantization6.75 GiB3.30 GiB
Q4_K_M18.14 GiB4.64 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextOlmo-3.1-32B-Instructgemma-4-E4B-itRatio
4,0961.00 GiB0.12 GiB8.13×
8,1921.34 GiB0.18 GiB7.56×
16,3841.84 GiB0.29 GiB6.42×
32,7682.84 GiB0.51 GiB5.62×
65,5364.84 GiB0.94 GiB5.13×
131,0728.84 GiB1.82 GiB4.86×