Is Ternary-Bonsai-1.7B-unpacked or gemma-3-1b-it smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

gemma-3-1b-it — its KV cache at 32K context is 23.9x smaller, because of how its attention layers cache.

Model comparison

Ternary-Bonsai-1.7B-unpacked vs gemma-3-1b-it

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-3-1b-it's KV cache at 32K is 23.9× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Ternary-Bonsai-1.7B-unpacked	gemma-3-1b-it
Parameters	1.7B	1000M
Architecture	qwen3	gemma3
Layers	28	26
Native context	32,768	32,768
Mixture of experts	no	no
Quantizations published	2	28
Smallest quantization	0.43 GiB	0.52 GiB
Q4_K_M	—	0.75 GiB
Licence	apache-2.0	gemma

KV cache by context

the term that decides long-context viability

Context	Ternary-Bonsai-1.7B-unpacked	gemma-3-1b-it	Ratio
4,096	0.44 GiB	0.04 GiB	11.79×
8,192	0.88 GiB	0.05 GiB	16.59×
16,384	1.75 GiB	0.08 GiB	20.84×
32,768	3.50 GiB	0.15 GiB	23.89×
65,536	7.00 GiB	0.27 GiB	25.78×
131,072	14.00 GiB	0.52 GiB	26.85×