This is a very simple ComfyUI beginner friendly Seamless Fast Inpainting workflow that will work with the Z-Image Base All-in-One model (with baked in Text Encoder &amp; VAE) that work well with my 8GB AMD GPU or anything similar from Nvidia. But if you have a much weaker GPU (4GB or lower) you can find other low grade hardware supported GGUF models from these HuggingFace Repos -Jayn7's HuggingFace repo for Z-Image Base GGUF models ---------------------------------------------------------<a target="_blank" rel="ugc" href="https://huggingface.co/jayn7/Z-Image-GGUF/tree/main">https://huggingface.co/jayn7/Z-Image-GGUF/tree/main</a>Unsloth's QWEN3 GGUF Text Encoders for Z-Image Base GGUF models -------------------------------------------------------------------<a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/Qwen3-4B-GGUF/tree/main">https://huggingface.co/unsloth/Qwen3-4B-GGUF/tree/main</a>The newly released Z-Image Base model, developed by Tongyi-MAI (Alibaba's AI lab), is the non-distilled foundation of the Z-Image family, designed to serve as the "raw" starting point for the open-source community. Unlike Z-Image Turbo, which is a distilled version optimized for ultra-fast performance and sub-second inference using as few as 8 steps, the Base model is a full-weight checkpoint that requires more steps (typically 28 to 50) but offers significantly higher diversity in poses, compositions, and artistic styles. While Z-Image Turbo is specifically tuned for immediate photorealism and consumer-grade GPU efficiency, Z-Image Base is built for researchers and creators who need a more flexible model for fine-tuning, training LoRAs, or achieving complex prompt adherence that the faster, more "opinionated" Turbo variant might miss. Compared to other specialized versions like Z-Image Edit, which is fine-tuned specifically for instruction-based image-to-image tasks, the Base model provides the most neutral and unrefined data distribution, making it the superior choice for those looking to push the 6-billion parameter S3-DiT architecture into new, creative territories beyond just speed.You can download your necessary Z-Image Base ALL-in-One model and LORA used on this workflow from CivitAI (Details are mentioned below). Make sure you have latest enough ComfyUI installation and install any necessary nodes for this workflow using ComfyUI manager and place the correct files (AIO model, LORAs etc) in correct places. Also check out my other workflows, LORAs, toolsets, datasets for SD 1.5 + SDXL 1.0, Pony, Illustrious, WAN 2.1, WAN 2.2, MagicWAN Image v2, QWEN, HunyuanImage-2.1, HiDream, KREA, Chroma, AuraFlow, lumina 2, Z-Image Base &amp; Turbo, Flux.2 Klein and Flux.1 Dev. Feel free to toss some yellow Buzz on stuffs you like.How to use this -#1. Just select your desired Z-Image Base All-in-One model (or swap it with GGUF model files) file first and now#2. one or multiple matching LORA(s) from LORA Manager to start#3. then load your masked image to write your positive and negative prompts for output.#4. select how many images you want (Change the number besides the "Run" button)#5. select image sampling methods, CFG, steps etc. settings#6. finally press the run button to generate. That's it..** LORA usage for this workflow is optional you can use it without any LORAs, use with 1 or 2 or any other number of LORAs, too add new LORAs press the L button on top to lunch LORA Manager on a new tab find your LORA and if you want to use that LORA just click the upward Kite button. You can use the mask editor node for ComfyUI (masking works very well with it) or any of your favourite image editor for masking. But I would strongly recommend you to use ComfyUI's mask editor node because sometimes ComfyUI fails to use masked image created using external tool.Required Files==================### Download Link for Z-Image Base ALL-in-One Model (FP8) -------------------------------------------------------------<a target="_blank" rel="ugc" href="https://civarchive.com/models/2173571?modelVersionId=2637423">https://civarchive.com/models/2173571?modelVersionId=2637423</a>### Download Link for Z-Image-Aesthetic LORA (Base v1) -----------------------------------------------------------<a target="_blank" rel="ugc" href="https://civarchive.com/models/2214707?modelVersionId=2635939">https://civarchive.com/models/2214707?modelVersionId=2635939</a>

Files with SHA256: ed978707ac18ebf1cc6e64b51135cb6761a8d4c9ae1c98a3d2a08369e786646e