Is Qwen3-Coder-30B-A3B-Instruct or Huihui-GLM-4.7-Flash-abliterated-57B smaller?

Qwen3-Coder-30B-A3B-Instruct is smaller at Q4_K_M: 17.28 GiB against 31.37 GiB.

Which uses less memory at long context?

Qwen3-Coder-30B-A3B-Instruct — its KV cache at 32K context is 1.4x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct vs Huihui-GLM-4.7-Flash-abliterated-57B

At Q4_K_M, Qwen3-Coder-30B-A3B-Instruct is the smaller download — 18,556,689,568 bytes against 33,687,658,720. At long context the gap widens: Qwen3-Coder-30B-A3B-Instruct's KV cache at 32K is 1.4× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3-Coder-30B-A3B-Instruct	Huihui-GLM-4.7-Flash-abliterated-57B
Parameters	30.5B	57.3B
Architecture	qwen3moe	deepseek2
Layers	48	63
Native context	262,144	202,752
Mixture of experts	yes, 128 experts	yes, 64 experts
Quantizations published	46	36
Smallest quantization	7.46 GiB	10.79 GiB
Q4_K_M	17.28 GiB	31.37 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct	Huihui-GLM-4.7-Flash-abliterated-57B	Ratio
4,096	0.38 GiB	0.52 GiB	1.39×
8,192	0.75 GiB	1.05 GiB	1.39×
16,384	1.50 GiB	2.09 GiB	1.39×
32,768	3.00 GiB	4.18 GiB	1.39×
65,536	6.00 GiB	8.37 GiB	1.39×
131,072	12.00 GiB	16.73 GiB	1.39×