Model comparison

Phi-3.5-MoE-instruct vs gemma-4-12B-it-qat-q4_0-unquantized

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

Phi-3.5-MoE-instructgemma-4-12B-it-qat-q4_0-unquantized
Parameters41.9B12.0B
Architecturephimoegemma4
Layers3248
Native context131,072262,144
Mixture of expertsyes, 16 expertsno
Quantizations published292
Smallest quantization10.27 GiB6.50 GiB
Q4_K_M23.61 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextPhi-3.5-MoE-instructgemma-4-12B-it-qat-q4_0-unquantizedRatio
4,0960.72 GiB
8,1920.97 GiB
16,3841.47 GiB
32,7682.47 GiB
65,5364.47 GiB
131,0728.47 GiB