Is Qwen2.5-Coder-1.5B-Instruct-abliterated or Qwen3-1.7B smaller?

Qwen3-1.7B is smaller at Q4_K_M: 1.03 GiB against 1.04 GiB.

Which uses less memory at long context?

Qwen2.5-Coder-1.5B-Instruct-abliterated — its KV cache at 32K context is 4.0x smaller, because of how its attention layers cache.

Model comparison

Qwen2.5-Coder-1.5B-Instruct-abliterated vs Qwen3-1.7B

At Q4_K_M, Qwen3-1.7B is the smaller download — 1,107,409,472 bytes against 1,117,321,696. At long context the gap widens: Qwen2.5-Coder-1.5B-Instruct-abliterated's KV cache at 32K is 4.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen2.5-Coder-1.5B-Instruct-abliterated	Qwen3-1.7B
Parameters	1.8B	2.0B
Architecture	qwen2	qwen3
Layers	28	28
Native context	32,768	40,960
Mixture of experts	no	no
Quantizations published	35	48
Smallest quantization	0.65 GiB	0.50 GiB
Q4_K_M	1.04 GiB	1.03 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	Qwen2.5-Coder-1.5B-Instruct-abliterated	Qwen3-1.7B	Ratio
4,096	0.11 GiB	0.44 GiB	4.00×
8,192	0.22 GiB	0.88 GiB	4.00×
16,384	0.44 GiB	1.75 GiB	4.00×
32,768	0.88 GiB	3.50 GiB	4.00×
65,536	1.75 GiB	7.00 GiB	4.00×
131,072	3.50 GiB	14.00 GiB	4.00×