Is granite-3.1-2b-instruct or Llama-3.2-1B-Instruct smaller?

Llama-3.2-1B-Instruct is smaller at Q4_K_M: 0.75 GiB against 1.43 GiB.

Which uses less memory at long context?

Llama-3.2-1B-Instruct — its KV cache at 32K context is 2.5x smaller, because of how its attention layers cache.

Model comparison

granite-3.1-2b-instruct vs Llama-3.2-1B-Instruct

At Q4_K_M, Llama-3.2-1B-Instruct is the smaller download — 807,690,688 bytes against 1,530,551,136. At long context the gap widens: Llama-3.2-1B-Instruct's KV cache at 32K is 2.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	granite-3.1-2b-instruct	Llama-3.2-1B-Instruct
Parameters	2.5B	1.2B
Architecture	granite	llama
Layers	40	16
Native context	131,072	131,072
Mixture of experts	no	no
Quantizations published	44	39
Smallest quantization	0.56 GiB	0.39 GiB
Q4_K_M	1.43 GiB	0.75 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	granite-3.1-2b-instruct	Llama-3.2-1B-Instruct	Ratio
4,096	0.31 GiB	0.13 GiB	2.50×
8,192	0.63 GiB	0.25 GiB	2.50×
16,384	1.25 GiB	0.50 GiB	2.50×
32,768	2.50 GiB	1.00 GiB	2.50×
65,536	5.00 GiB	2.00 GiB	2.50×
131,072	10.00 GiB	4.00 GiB	2.50×