Model comparison

GigaAM-v3 vs canary-180m-flash

At Q4_K_M, canary-180m-flash is the smaller download — 139,223,744 bytes against 182,497,888.

From the file· summed bytes, KV per layer

Side by side

GigaAM-v3canary-180m-flash
Parameters223M189M
Architecturegigaamcanary
Layers
Native context
Mixture of expertsnono
Quantizations published186
Smallest quantization0.17 GiB0.13 GiB
Q4_K_M0.17 GiB0.13 GiB
Licencemitcc-by-4.0

KV cache by context

the term that decides long-context viability
ContextGigaAM-v3canary-180m-flashRatio
4,096
8,192
16,384
32,768
65,536
131,072