[SD 1.5][Checkpoint]Elegance Elegance 3.7
<h1>Read Description</h1><p></p><p><strong><u>VAE not required but recommended</u></strong></p><p>VAE is recommended - <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main">https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main</a></p><p></p><p>File Structure for AUTOMATIC1111-webui :</p><p>|──sd</p><p>|____|──stable-diffusion-webui</p><p>|____|____|──models</p><p>|____|____|____|──VAE</p><p>|____|____|____|____|──<Put your VAE file here></p><p></p><p><strong><u>Merged Models</u></strong></p><p>A list of merged models can be found bellow in the description of the attached model version.</p><p></p><p>This model used an unconventional method of merging, more information available in the models description.</p><p></p><p><strong>This model is experimental</strong></p><p></p><p><strong><u>Capabilities</u></strong></p><ul><li><p>NSFW/SFW portraits</p></li><li><p>Photorealistic 3D renders</p></li><li><p>Landscape and Architecture photography/renders</p></li><li><p>Emphasis on photorealism and detail</p></li></ul><p><strong><u>Limitations</u></strong></p><p>Anything not listed above. I'm unsure of how this model will handle stylized images, admittedly I'm not great at writing stylized prompts. Please do test and share your findings!</p><p></p><p></p><p></p><p><strong><u>Trigger Words</u></strong></p><p>I'm not aware of any trigger words that have drastic influence on the generation process.</p><p>However, tags such as:</p><p><code>"3d render", "cartoon" | "nsfw", "sfw", "nudity", and "erotica"</code></p><p>tend to add push the generation (to some degree) in one direction or another. For example, putting <code>sfw</code> in your prompt and <code>nsfw</code> in your negative prompt <em>should</em> push the generation to produce a SFW image.</p><p></p><p></p><p><strong><u>Prompt Guide</u></strong></p><p>UnstablePhotorealv.5 was used in this merge, as a result comma separated tags <em>can</em> achieve decent results. Although I recommend using a combination of separated commas + the natural language flow you're likely more familiar with.</p><p>Refer to this document for more information on UnstablePhotoreal tagging - <a target="_blank" rel="ugc" href="https://docs.google.com/document/d/1-DDIHVbsYfynTp_rsKLu4b2tSQgxtO5F6pNsNla12k0/edit#heading=h.3znysh7">https://docs.google.com/document/d/1-DDIHVbsYfynTp_rsKLu4b2tSQgxtO5F6pNsNla12k0/edit#heading=h.3znysh7</a></p><p></p><p></p><p><strong><u>What's Next?</u></strong></p><ul><li><p>Fine-tuned dreambooth models that focus on areas current SD models fall short.</p></li><li><p>Hypernetwork, or tuned decoder for better handheld object holding.</p></li><li><p>Fine-tuned dreambooth model for firearms and various other weaponry. (This ties in with the aforementioned Hypernetwork).</p></li><li><p>R&D for LoRA.</p></li><li><p>R&D for new methods of merging.</p></li><li><p>Miscellaneous scripts and utilities that I create while working on SD-related projects.</p></li><li><p>Text-to-inpainting extension for AUTOMATIC1111 webui</p></li></ul><p></p><p><strong><u>Link(s) to my other models</u></strong></p><ul><li><p>Refined - <a target="_blank" rel="ugc" href="https://civitai.com/models/8392/refined">https://civitai.com/models/8392/refined</a></p></li><li><p>Experience - <a target="_blank" rel="ugc" href="https://civitai.com/models/5952/experience">https://civitai.com/models/5952/experience</a></p></li></ul><ul><li><p>Clarity - <a target="_blank" rel="ugc" href="https://civitai.com/models/5062/clarity">https://civitai.com/models/5062/clarity</a></p></li><li><p>VisionGen - Realism -<a target="_blank" rel="ugc" href="https://civitai.com/models/4834/visiongen-realism">https://civitai.com/models/4834/visiongen-realism</a></p></li></ul>
页:
[1]