Model comparison

SeedVR2-3B vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, SeedVR2-3B is the smaller download — 11,511,957,408 bytes against 18,556,685,824.

From the file· summed bytes, KV per layer

Side by side

SeedVR2-3BQwen3-30B-A3B-Thinking-2507
Parameters3.4B30.5B
Architectureseedvrqwen3moe
Layers48
Native context262,144
Mixture of expertsnoyes, 128 experts
Quantizations published251
Smallest quantization3.41 GiB7.05 GiB
Q4_K_M10.72 GiB17.28 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextSeedVR2-3BQwen3-30B-A3B-Thinking-2507Ratio
4,0960.38 GiB
8,1920.75 GiB
16,3841.50 GiB
32,7683.00 GiB
65,5366.00 GiB
131,07212.00 GiB