Is Qwen3-VL-Embedding-2B or Qwen3-VL-4B-Instruct smaller?

Qwen3-VL-Embedding-2B is smaller at Q4_K_M: 1.03 GiB against 2.33 GiB.

Which uses less memory at long context?

Qwen3-VL-Embedding-2B — its KV cache at 32K context is 1.3x smaller, because of how its attention layers cache.

Model comparison

Qwen3-VL-Embedding-2B vs Qwen3-VL-4B-Instruct

At Q4_K_M, Qwen3-VL-Embedding-2B is the smaller download — 1,107,410,528 bytes against 2,497,281,568. At long context the gap widens: Qwen3-VL-Embedding-2B's KV cache at 32K is 1.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3-VL-Embedding-2B	Qwen3-VL-4B-Instruct
Parameters	2.1B	4.4B
Architecture	qwen3vl	qwen3vl
Layers	28	36
Native context	262,144	262,144
Mixture of experts	no	no
Quantizations published	11	26
Smallest quantization	0.72 GiB	1.01 GiB
Q4_K_M	1.03 GiB	2.33 GiB
Licence	—	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3-VL-Embedding-2B	Qwen3-VL-4B-Instruct	Ratio
4,096	0.44 GiB	0.56 GiB	1.29×
8,192	0.88 GiB	1.13 GiB	1.29×
16,384	1.75 GiB	2.25 GiB	1.29×
32,768	3.50 GiB	4.50 GiB	1.29×
65,536	7.00 GiB	9.00 GiB	1.29×
131,072	14.00 GiB	18.00 GiB	1.29×