Is stable-code-3b or Qwen3-1.7B smaller?

Qwen3-1.7B is smaller at Q4_K_M: 1.03 GiB against 1.59 GiB.

Which uses less memory at long context?

Qwen3-1.7B — its KV cache at 32K context is 2.9x smaller, because of how its attention layers cache.

Model comparison

stable-code-3b vs Qwen3-1.7B

At Q4_K_M, Qwen3-1.7B is the smaller download — 1,107,409,472 bytes against 1,708,595,200. At long context the gap widens: Qwen3-1.7B's KV cache at 32K is 2.9× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	stable-code-3b	Qwen3-1.7B
Parameters	2.8B	2.0B
Architecture	stablelm	qwen3
Layers	32	28
Native context	16,384	40,960
Mixture of experts	no	no
Quantizations published	48	48
Smallest quantization	0.63 GiB	0.50 GiB
Q4_K_M	1.59 GiB	1.03 GiB
Licence	other	—

KV cache by context

the term that decides long-context viability

Context	stable-code-3b	Qwen3-1.7B	Ratio
4,096	1.25 GiB	0.44 GiB	2.86×
8,192	2.50 GiB	0.88 GiB	2.86×
16,384	5.00 GiB	1.75 GiB	2.86×
32,768	10.00 GiB	3.50 GiB	2.86×
65,536	20.00 GiB	7.00 GiB	2.86×
131,072	40.00 GiB	14.00 GiB	2.86×