Uncensored KREA2 Examples images shows the base model (int8) on the left / finetune (int8) on the right (same seed same prompt, 8 steps) This V3 was retrained, meant to be semi-realistic, have less aesthetic than V2, but finer details, was trained at all available buckets from 512 to 2048 px, with a different methodology than V2. Then was blocks filtered, tweaked with a tool I have made that refine blocks weights by manual iterations over multiple rounds of A/B comparison discrimination. You can use my "Magic LoRA Loader" from <a target="_blank" rel="ugc" href="https://github.com/PGCRT/CRT-Nodes">https://github.com/PGCRT/CRT-Nodes</a> If you want inference blocks tweaks Tip - For sharper output, you can use the <a target="_blank" rel="ugc" href="https://huggingface.co/spacepxl/Wan2.1-VAE-upscale2x/blob/main/Wan2.1_VAE_upscale2x_imageonly_real_v1.safetensors"><span style="background-color:rgba(255, 255, 255, 0.03);color:rgb(231, 233, 234);font-family:TwitterChirp, -apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif;font-size:15px">Wan2.1_VAE_upscale2x_imageonly_real_v1</a> with <a target="_blank" rel="ugc" href="https://github.com/spacepxl/ComfyUI-VAE-Utils">https://github.com/spacepxl/ComfyUI-VAE-Utils</a> Loader / decoder Get free credit between 5-500$ in runpod with my link <a target="_blank" rel="ugc" href="https://runpod.io?ref=u7b2habt">https://runpod.io?ref=u7b2habt</a> <a rel="ugc" href="https://buymeacoffee.com/designedbycrt">Or buy me a coffee</a> If you want to support my work. Have fun

Files with SHA256: f4583174ad3bc00b166191454cc3cba6d4c191580aa0287026a8da98f61ce680