RORem mixed needs about 2.8 GB of memory at Q4_K (a 2.0 GB file) with an 8K context window, so it fits on a 4 GB GPU.
RORem-mixed Q4_K GGUF for SDXL inpainting UNet
RORem-mixed is a 2.57B parameter SDXL inpainting UNet converted to Q4_K GGUF. It supports 512x512 and 1024x1024 inputs and is designed for object removal and inpainting tasks. The model is not a full SDXL pipeline and requires external VAE, CLIP-L, and CLIP-G weights.
Good at
- Object removal in 512x512 and 1024x1024
- Inpainting with mask dilation
- Stable diffusion inpainting with Q4_K quantization
Limits
- Not a complete SDXL pipeline
- Requires external VAE and CLIP weights
- No general text-to-image generation
Best for
- Manga inpainting
- Object removal in images
- Mask-based editing
Who it’s for. Developers and artists who need a lightweight inpainting model for SDXL workflows. The model is suitable for local inference with stable‑diffusion.cpp and requires a CUDA build.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 2.0 GB | 2.8 GB | Fits | Get file → |
The Q4_K quantization file is 2.02 GB and needs about 2.8 GB of VRAM at 8 k context. For 8 GB VRAM, Q4_K is the only available quant. If more VRAM is available, consider higher precision quantizations if they exist.
What GPU do you need for RORem mixed?
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 (2.0 GB file, about 2.8 GB in use).
Which RORem mixed quant should you choose?
Q4_K (2.0 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about RORem mixed
How much VRAM does RORem mixed need?
About 2.8 GB at Q4_K 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 RORem mixed on an 8 GB GPU?
Yes, with Q4_K: the 2.0 GB file needs about 2.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of RORem mixed GGUF?
It depends on the quant: Q4_K is 2.0 GB.
Where can I download RORem mixed GGUF?
The files are on Hugging Face in the mayocream/RORem-mixed-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does RORem mixed use?
Hugging Face lists the license as openrail++. 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 RORem-mixed on 8 GB VRAM?
Yes, the Q4_K quantization requires approximately 2.8 GB of VRAM at 8 k context, which fits within 8 GB.
Does RORem-mixed support text-to-image generation?
No, it is intended only for inpainting/object removal and does not support general text-to-image generation.
What additional weights are needed for a full SDXL pipeline?
You need the VAE, CLIP-L, and CLIP-G weights from diffusers/stable-diffusion-xl-1.0-inpainting-0.1.