Model comparison

SmolLM-1.7B-Instruct-v0.2 vs gemma-3-1b-it

At Q4_K_M, gemma-3-1b-it is the smaller download — 806,058,240 bytes against 1,055,610,144.

From the file· summed bytes, KV per layer

Side by side

SmolLM-1.7B-Instruct-v0.2gemma-3-1b-it
Parameters1.7B1000M
Architecturellamagemma3
Layers26
Native context32,768
Mixture of expertsnono
Quantizations published3228
Smallest quantization0.38 GiB0.52 GiB
Q4_K_M0.98 GiB0.75 GiB
Licenceapache-2.0gemma

KV cache by context

the term that decides long-context viability
ContextSmolLM-1.7B-Instruct-v0.2gemma-3-1b-itRatio
4,0960.04 GiB
8,1920.05 GiB
16,3840.08 GiB
32,7680.15 GiB
65,5360.27 GiB
131,0720.52 GiB
SmolLM-1.7B-Instruct-v0.2 vs gemma-3-1b-it — size, memory and hardware fit — ossmodeldb