Is Qwen2.5-Coder-1.5B-Instruct-abliterated or HyperCLOVAX-SEED-Text-Instruct-1.5B smaller?

Qwen2.5-Coder-1.5B-Instruct-abliterated is smaller at Q4_K_M: 1.04 GiB against 1.06 GiB.

Which uses less memory at long context?

We cannot resolve the KV layout for one of these models, so we do not publish a comparison.

Model comparison

At Q4_K_M, Qwen2.5-Coder-1.5B-Instruct-abliterated is the smaller download — 1,117,321,696 bytes against 1,133,974,368.

From the file· summed bytes, KV per layer

	Qwen2.5-Coder-1.5B-Instruct-abliterated	HyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters	1.8B	1.6B
Architecture	qwen2	llama
Layers	28	—
Native context	32,768	—
Mixture of experts	no	no
Quantizations published	35	1
Smallest quantization	0.65 GiB	1.06 GiB
Q4_K_M	1.04 GiB	1.06 GiB
Licence	apache-2.0	other

the term that decides long-context viability

Context	Qwen2.5-Coder-1.5B-Instruct-abliterated	HyperCLOVAX-SEED-Text-Instruct-1.5B	Ratio
4,096	0.11 GiB	—	—
8,192	0.22 GiB	—	—
16,384	0.44 GiB	—	—
32,768	0.88 GiB	—	—
65,536	1.75 GiB	—	—
131,072	3.50 GiB	—	—