SeedVR2-3B · AInVFX

SeedVR2-3B Q4_K_M

This file is exactly 11,511,957,408 bytes — 10.72 GiB / 11.51 GB at an effective 27.155 bits per weight. The nominal rate for Q4_K_M is lower; mixed-precision tensors make the real figure higher, always.

From the file· 3 file(s), summed

Get it

10.72 GiB · 3 files
llama.cpp
llama-cli -hf AInVFX/SeedVR2_comfyUI:Q4_K_M

Downloads and runs in one step, resolving the quantization by name.

Hugging Face CLI
hf download AInVFX/SeedVR2_comfyUI --include "seedvr2_ema_3b-Q4_K_M.gguf"

This quantization is split across 3 files totalling 10.72 GiB. You need every part — the first is not a smaller version of the model.

Direct downloads — all 3 required

Straight from the Hugging Face CDN — we host nothing and earn nothing from this. Verify what you received against the exact byte count above; a size mismatch is the usual cause of a file that will not load.

Repository →
Size
10.72 GiB
11.51 GB
Effective bpw
27.155
from real bytes ÷ params
Tensors
—
Header
—
GGUF metadata
SeedVR2-3B Q4_K_M — exact size and VRAM — ossmodeldb