SeedVR2 models needs about 2.6 GB of memory at Q4_K_M (a 1.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.9 GB (Q4_K_M) to 4.6 GB (Q4_K_M).
SeedVR2 3B video super‑resolution model
SeedVR2 is a diffusion‑based video super‑resolution model. It upsamples low‑resolution video to high resolution. The model is available in a 3.39B parameter variant.
Good at
- Video upscaling
- Diffusion‑based restoration
- High‑quality output
Limits
- No vision support beyond video upscaling
- Limited to 8k context
- Apache‑2.0 license
Best for
- Enhancing low‑res footage
- Post‑production upscaling
- AI video restoration
Who it’s for. The model is suited for developers and content creators who need high‑quality video upscaling. It requires a GPU with at least 8GB VRAM and supports the ComfyUI‑FFMPEGA extension.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 1.9 GB | 2.6 GB | Fits | Get file → | |
| Q4_K_M | 4.6 GB | 5.7 GB | Fits | Get file → |
For 8GB VRAM, the Q4_K_M 1.91GB file uses about 2.6GB memory at 8k context, which fits comfortably. If more memory is available, the 4.64GB Q4_K_M file uses about 5.7GB memory and provides the same quality. Both are approximate estimates.
What GPU do you need for SeedVR2 models?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q4_K_M (1.9 GB file, about 2.6 GB in use).
- 6 GB VRAM: the best quant that fits is Q4_K_M (4.6 GB file, about 5.7 GB in use).
Which SeedVR2 models quant should you choose?
Q4_K_M is the usual starting point for SeedVR2 models: it balances file size and quality well. Here it is a 1.9 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SeedVR2 models
How much VRAM does SeedVR2 models need?
About 2.6 GB at Q4_K_M with an 8K context window. That is the 1.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run SeedVR2 models on an 8 GB GPU?
Yes, with Q4_K_M: the 4.6 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of SeedVR2 models GGUF?
It depends on the quant: Q4_K_M is 1.9 GB, Q4_K_M is 4.6 GB.
Where can I download SeedVR2 models GGUF?
The files are on Hugging Face in the AEmotionStudio/SeedVR2-models repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does SeedVR2 models use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run SeedVR2 on 8GB VRAM?
Yes, the 1.91GB Q4_K_M quant uses about 2.6GB memory at 8k context, fitting within 8GB VRAM.
What license does SeedVR2 use?
SeedVR2 is released under the Apache‑2.0 license.
Does SeedVR2 support other tasks besides video upscaling?
No, SeedVR2 is specifically designed for video super‑resolution and upscaling.