Is Qwen3-Coder-Next-DFlash or Qwen3-0.6B smaller?

Qwen3-Coder-Next-DFlash is smaller at Q4_K_M: 0.27 GiB against 0.37 GiB.

Which uses less memory at long context?

Qwen3-Coder-Next-DFlash — its KV cache at 32K context is 7.0x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-Next-DFlash vs Qwen3-0.6B

At Q4_K_M, Qwen3-Coder-Next-DFlash is the smaller download — 286,754,016 bytes against 396,705,472. At long context the gap widens: Qwen3-Coder-Next-DFlash's KV cache at 32K is 7.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3-Coder-Next-DFlash	Qwen3-0.6B
Parameters	474M	752M
Architecture	dflash	qwen3
Layers	8	28
Native context	262,144	40,960
Mixture of experts	no	no
Quantizations published	7	44
Smallest quantization	0.25 GiB	0.20 GiB
Q4_K_M	0.27 GiB	0.37 GiB
Licence	—	—

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-Next-DFlash	Qwen3-0.6B	Ratio
4,096	0.06 GiB	0.44 GiB	7.00×
8,192	0.13 GiB	0.88 GiB	7.00×
16,384	0.25 GiB	1.75 GiB	7.00×
32,768	0.50 GiB	3.50 GiB	7.00×
65,536	1.00 GiB	7.00 GiB	7.00×
131,072	2.00 GiB	14.00 GiB	7.00×