SeedVR2 comfyUI needs about 2.7 GB of memory at Q4_K_M (a 2.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 2.0 GB (Q4_K_M) to 4.8 GB (Q4_K_M).
SeedVR2 comfyUI – one‑step diffusion video restoration model
SeedVR2 comfyUI is a 7.8‑billion parameter diffusion model for video restoration and upscaling. It is optimized for ComfyUI and supports single‑step inference. The model is released under Apache‑2.0 and is available in several GGUF quantizations.
Good at
- High‑resolution video upscaling
- Temporal consistency in restored frames
- One‑step inference
Limits
- No text generation support
- Requires GPU with at least 8 GB VRAM for Q4_K_M 2 GB variant
- Performance may degrade on heavy motion or extreme degradations
Best for
- Restoring low‑quality surveillance footage
- Enhancing 720p AIGC videos
- Batch video upscaling in ComfyUI
Who it’s for. Users needing fast, high‑quality video restoration in a ComfyUI workflow. Suitable for researchers and developers working with diffusion‑based video models.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 2.0 GB | 2.7 GB | Fits | Get file → | |
| Q8_0 | 3.7 GB | 4.6 GB | Fits | Get file → | |
| Q4_K_M | 4.8 GB | 5.8 GB | Fits | Get file → | |
| Q4_K_M | 4.8 GB | 5.8 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M 2 GB (est. 2.7 GB memory). For 12 GB VRAM, Q8_0 3.66 GB (est. 4.6 GB). For 16 GB VRAM, Q4_K_M 4.76 GB (est. 5.8 GB). Memory estimates are approximate.
What GPU do you need for SeedVR2 comfyUI?
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 (2.0 GB file, about 2.7 GB in use).
- 6 GB VRAM: the best quant that fits is Q4_K_M (4.8 GB file, about 5.8 GB in use).
Which SeedVR2 comfyUI quant should you choose?
Q4_K_M is the usual starting point for SeedVR2 comfyUI: it balances file size and quality well. Here it is a 2.0 GB file.
The Q8_0 file is 3.7 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SeedVR2 comfyUI
How much VRAM does SeedVR2 comfyUI need?
About 2.7 GB at Q4_K_M with an 8K context window. That is the 2.0 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 comfyUI on an 8 GB GPU?
Yes, with Q4_K_M: the 4.8 GB file needs about 5.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of SeedVR2 comfyUI GGUF?
It depends on the quant: Q4_K_M is 2.0 GB, Q8_0 is 3.7 GB, Q4_K_M is 4.8 GB, Q4_K_M is 4.8 GB.
Where can I download SeedVR2 comfyUI GGUF?
The files are on Hugging Face in the AInVFX/SeedVR2_comfyUI 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 comfyUI 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 8 GB VRAM?
Yes, the Q4_K_M 2 GB quantization requires about 2.7 GB VRAM at 8 k context, which fits within 8 GB.
Does SeedVR2 support text prompts?
No, SeedVR2 is a video restoration model and does not handle text generation.
What license applies to SeedVR2?
SeedVR2 is released under the Apache‑2.0 license.