Is Qwen3.5-9B-DFlash or jina-embeddings-v5-text-small smaller?

jina-embeddings-v5-text-small is smaller at Q4_K_M: 0.37 GiB against 0.71 GiB.

Which uses less memory at long context?

Qwen3.5-9B-DFlash — its KV cache at 32K context is 16.4x smaller, because of how its attention layers cache.

Model comparison

Qwen3.5-9B-DFlash vs jina-embeddings-v5-text-small

At Q4_K_M, jina-embeddings-v5-text-small is the smaller download — 396,705,152 bytes against 765,959,872. At long context the gap widens: Qwen3.5-9B-DFlash's KV cache at 32K is 16.4× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.5-9B-DFlash	jina-embeddings-v5-text-small
Parameters	1.3B	596M
Architecture	dflash	qwen3
Layers	6	28
Native context	262,144	32,768
Mixture of experts	no	no
Quantizations published	11	56
Smallest quantization	0.45 GiB	0.19 GiB
Q4_K_M	0.71 GiB	0.37 GiB
Licence	—	cc-by-nc-4.0

KV cache by context

the term that decides long-context viability

Context	Qwen3.5-9B-DFlash	jina-embeddings-v5-text-small	Ratio
4,096	0.09 GiB	0.44 GiB	4.67×
8,192	0.12 GiB	0.88 GiB	7.34×
16,384	0.15 GiB	1.75 GiB	11.64×
32,768	0.21 GiB	3.50 GiB	16.44×
65,536	0.34 GiB	7.00 GiB	20.72×
131,072	0.59 GiB	14.00 GiB	23.81×