Is Falcon3-1B-Instruct or Llama-3.2-1B-Instruct smaller?

Llama-3.2-1B-Instruct is smaller at Q4_K_M: 0.75 GiB against 0.98 GiB.

Which uses less memory at long context?

Llama-3.2-1B-Instruct — its KV cache at 32K context is 2.3x smaller, because of how its attention layers cache.

Model comparison

Falcon3-1B-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 1,057,044,608. At long context the gap widens: Llama-3.2-1B-Instruct's KV cache at 32K is 2.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Falcon3-1B-Instruct	Llama-3.2-1B-Instruct
Parameters	1.7B	1.2B
Architecture	llama	llama
Layers	18	16
Native context	8,192	131,072
Mixture of experts	no	no
Quantizations published	30	39
Smallest quantization	0.21 GiB	0.39 GiB
Q4_K_M	0.98 GiB	0.75 GiB
Licence	other	—

KV cache by context

the term that decides long-context viability

Context	Falcon3-1B-Instruct	Llama-3.2-1B-Instruct	Ratio
4,096	0.28 GiB	0.13 GiB	2.25×
8,192	0.56 GiB	0.25 GiB	2.25×
16,384	1.13 GiB	0.50 GiB	2.25×
32,768	2.25 GiB	1.00 GiB	2.25×
65,536	4.50 GiB	2.00 GiB	2.25×
131,072	9.00 GiB	4.00 GiB	2.25×