Is LFM2.5-350M or Llama-3.2-1B-Instruct smaller?

LFM2.5-350M is smaller at Q4_K_M: 0.21 GiB against 0.75 GiB.

Which uses less memory at long context?

LFM2.5-350M — its KV cache at 32K context is 2.7x smaller, because of how its attention layers cache.

Model comparison

LFM2.5-350M vs Llama-3.2-1B-Instruct

At Q4_K_M, LFM2.5-350M is the smaller download — 228,332,128 bytes against 807,690,688. At long context the gap widens: LFM2.5-350M's KV cache at 32K is 2.7× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	LFM2.5-350M	Llama-3.2-1B-Instruct
Parameters	354M	1.2B
Architecture	lfm2	llama
Layers	16	16
Native context	128,000	131,072
Mixture of experts	no	no
Quantizations published	13	39
Smallest quantization	0.20 GiB	0.39 GiB
Q4_K_M	0.21 GiB	0.75 GiB
Licence	other	—

KV cache by context

the term that decides long-context viability

Context	LFM2.5-350M	Llama-3.2-1B-Instruct	Ratio
4,096	0.05 GiB	0.13 GiB	2.67×
8,192	0.09 GiB	0.25 GiB	2.67×
16,384	0.19 GiB	0.50 GiB	2.67×
32,768	0.38 GiB	1.00 GiB	2.67×
65,536	0.75 GiB	2.00 GiB	2.67×
131,072	1.50 GiB	4.00 GiB	2.67×