Is Qwen3-Coder-30B-A3B-Instruct or openbuddy-deepseek-67b-v15.3-4k smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

Qwen3-Coder-30B-A3B-Instruct — its KV cache at 32K context is 4.0x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct vs openbuddy-deepseek-67b-v15.3-4k

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: Qwen3-Coder-30B-A3B-Instruct'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

	Qwen3-Coder-30B-A3B-Instruct	openbuddy-deepseek-67b-v15.3-4k
Parameters	30.5B	67.4B
Architecture	qwen3moe	llama
Layers	48	95
Native context	262,144	4,096
Mixture of experts	yes, 128 experts	no
Quantizations published	46	20
Smallest quantization	7.46 GiB	13.65 GiB
Q4_K_M	17.28 GiB	—
Licence	apache-2.0	other

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct	openbuddy-deepseek-67b-v15.3-4k	Ratio
4,096	0.38 GiB	1.48 GiB	3.96×
8,192	0.75 GiB	2.97 GiB	3.96×
16,384	1.50 GiB	5.94 GiB	3.96×
32,768	3.00 GiB	11.88 GiB	3.96×
65,536	6.00 GiB	23.75 GiB	3.96×
131,072	12.00 GiB	47.50 GiB	3.96×