Is next-1b or Qwen3-4B smaller?

next-1b is smaller at Q4_K_M: 0.75 GiB against 2.33 GiB.

Which uses less memory at long context?

next-1b — its KV cache at 32K context is 30.7x smaller, because of how its attention layers cache.

Model comparison

next-1b vs Qwen3-4B

At Q4_K_M, next-1b is the smaller download — 806,057,664 bytes against 2,497,280,256. At long context the gap widens: next-1b's KV cache at 32K is 30.7× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	next-1b	Qwen3-4B
Parameters	1000M	4.0B
Architecture	gemma3	qwen3
Layers	26	36
Native context	32,768	40,960
Mixture of experts	no	no
Quantizations published	1	29
Smallest quantization	0.75 GiB	1.01 GiB
Q4_K_M	0.75 GiB	2.33 GiB
Licence	mit	—

KV cache by context

the term that decides long-context viability

Context	next-1b	Qwen3-4B	Ratio
4,096	0.04 GiB	0.56 GiB	15.16×
8,192	0.05 GiB	1.13 GiB	21.33×
16,384	0.08 GiB	2.25 GiB	26.79×
32,768	0.15 GiB	4.50 GiB	30.72×
65,536	0.27 GiB	9.00 GiB	33.15×
131,072	0.52 GiB	18.00 GiB	34.52×