Model comparison

Unlimited-OCR vs gemma-4-E2B-it

At Q4_K_M, Unlimited-OCR is the smaller download — 1,950,321,792 bytes against 3,106,738,272.

From the file· summed bytes, KV per layer

Side by side

Unlimited-OCRgemma-4-E2B-it
Parameters3.3B5.1B
Architecturedeepseek2-ocrgemma4
Layers1235
Native context32,768131,072
Mixture of expertsyes, 64 expertsno
Quantizations published2154
Smallest quantization1.15 GiB2.13 GiB
Q4_K_M1.82 GiB2.89 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextUnlimited-OCRgemma-4-E2B-itRatio
4,0960.05 GiB
8,1920.08 GiB
16,3840.14 GiB
32,7680.25 GiB
65,5360.46 GiB
131,0720.90 GiB