RealVis Medium - Model for Realistic and Photorealistic Image Generation

Overview

RealVis Medium is a fine-tuned model with a focus on generating high-quality realistic and photorealistic images, based on SD3.5 Medium.

Warning: This is not the final version of the model. Work on improving the model is still ongoing.

You can support me directly on my Boosty

For ComfyUI

Download all necessary model files and move them to the following paths:

From text_encoders folder, the files clip_g.safetensors, clip_l.safetensors, t5xxl_fp16.safetensors, or t5xxl_fp8_e4m3fn.safetensors to the location ComfyUI\models\clip
From transformer folder, the file diffusion_pytorch_model.safetensors to the location ComfyUI\models\unet
From vae folder, the file diffusion_pytorch_model.safetensors to the location ComfyUI\models\vae

While this model can handle long prompts, you may observe artifacts on the edge of generations when T5 tokens go over 256. Pay attention to the token limits when using this model in your workflow, and shortern prompts if artifacts becomes too obvious.
The medium model has a different training data distribution than the large model, so it may not respond to the same prompt similarly.
We recommend sampling with Skip Layer Guidance for better structure and anatomy coherency.