FireRed-Image-Edit 1.1All In One Model version with built-in VAE and CLIP.Roadmap: Working on a new version with integrated LoRAs. I’d love to hear your thoughts—feel free to drop feedback or suggestions in the comments!.<ul><li>V1 : Using LoRa 8 Steps by FireRed Team</li><li>V2 : Mix LoRa Qwen Image 2512 : 4 Steps and 8 Steps, Good For Image Editing or Image Generation (If at high resolution when doing Text to Image you do not use pixels that are multiples of 32, then the model will create strange character duplications.)</li><li>I'm experimenting with an ablation of the base model. This model is quite rigid—or at the very least, it demands more descriptive prompts.For example: if you say "make an NSFW image," the model triggers the censor.But if you use a prompt like: "create an artistic image [with specific NSFW details...]", the model will actually fulfill the request.</li></ul><hr />Reccomended SettingsBase-Version :<ul><li>CFG : 4</li><li>Steps 20-30</li><li>Sampler : Euler_Ancestral</li></ul>FLASH-Version :<ul><li>CFG : 1</li><li>Steps : 4-8</li><li>Sampler : Euler_Ancestral</li></ul>Base Model: <a target="_blank" rel="ugc" href="https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.1">https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.1</a><hr />Support &amp; Custom Requests: <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Processing, testing, and optimizing these models require significant computational resources and time. If you require a specific Checkpoint or LoRA to be optimized for your hardware constraints, you may submit a formal request and support this ongoing work through the link below: <a target="_blank" rel="ugc" href="https://ko-fi.com/morikomorizz">https://ko-fi.com/morikomorizz</a>

FireRed-FLASH-AIO

<ul><li>Only Need 4 Steps for fast, 8 Steps for Better Quality</li><li>Combined With My Qwen-Flash64 LoRa Steps</li></ul>

Files with SHA256: 3b96e30999e80a70525f32128db376976f7a798f56ea729f063341636f934725