<h1 id="this-is-a-workflow-for-moss-imageanimasdxl(noobai-xlillustrious-xl).">This is a workflow for MOSS Image/Anima/SDXL(NoobAI XL/Illustrious-XL).</h1><h3 id="last-update:-moss-image">Last update: MOSS Image</h3><h3 id="moss-image">MOSS Image</h3>This is an example of a character profile, three-view drawing, and character coordinate specification using a bounding box (BBOX).I am using a customized version of Create Bounding Boxes.<a target="_blank" rel="ugc" href="https://github.com/hybskgks28275/ComfyUI-MossImageTools">ComfyUI-MossImageTools</a><h3 id="lora-merge-experimental">LoRA-Merge-EXPERIMENTAL</h3><ul><li>SSR-Merge</li></ul><a target="_blank" rel="ugc" href="https://arxiv.org/abs/2606.10617">SSR-Merge</a><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">SSR-Merge is intended for merging two character LoRAs, or a character LoRA with an outfit LoRA.When merging multiple character LoRAs to generate two characters simultaneously, using only Danbooru(Gelbooru) tags can easily cause the characters' features to become mixed. Therefore, please describe each character's distinctive features in natural language.<ul><li>NP-LoRA</li></ul><a target="_blank" rel="ugc" href="https://arxiv.org/abs/2511.11051">NP-LoRA</a><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">NP-LoRA is intended for adjusting and merging a character LoRA or concept LoRA while preserving the art style of the style LoRA.<h3 id="anima-xy-plot-lora">Anima XY Plot LoRA</h3><ul><li>XY Plot LoRA1</li></ul>This is an example of comparing results without LoRA applied, with LoRA applied, or between multiple LoRAs.<ul><li>XY Plot LoRA2</li></ul>This is an example of comparing different LoRA strength values.<h2 id="anima-turbo-lora-+-nag">Anima Turbo LoRA + NAG</h2>By using <a target="_blank" rel="ugc" href="https://civarchive.com/models/2560840/">Anima Turbo LoRA</a>, you can use Negative Prompts with NAG while maintaining low steps and CFG 1.Since applying NAG reduces generation speed, when generating with the 8–12 steps recommended for Anima Turbo LoRA, it is recommended to set it to approximately half of the total steps, such as start_percent=0 and end_percent=0.5.<h2 id="anima-couple">Anima Couple</h2><ul><li>v1</li></ul>The mask is split to draw the characters separately. If a single LoRA is trained on multiple characters, it can distinguish between them, but it does not support combining multiple character LoRAs.Since Anima generation takes time, this uses a speed-up node (Accelerate subgraph). In exchange for faster generation, some degradation may occur. Adjust the parameters as needed.<ul><li>v2</li></ul>Support for applying multiple character LoRAs has been added.<h2 id="anima">Anima</h2>The processing order is: generation → Hires.fix → face-only Hires.fix → hands-only Hires.fix → upscale.<h2 id="anima-xy-plot-checkpoint">Anima XY Plot CheckPoint</h2>This is a workflow for comparing Anima-based checkpoints.<h2 id="require">Require</h2><ul><li>SageAttention</li></ul><a target="_blank" rel="ugc" href="https://github.com/woct0rdho/SageAttention/releases">here</a>Please download and install the appropriate version for your environment.<ul><li>triton-windows</li></ul>ex)<pre><code>pip install triton-windows</code></pre><ul><li>sam3</li></ul><a target="_blank" rel="ugc" href="https://huggingface.co/facebook/sam3">v3.0 here</a><a target="_blank" rel="ugc" href="https://huggingface.co/facebook/sam3.1">v3.1 here</a><ul><li>my custom nodes</li></ul><a target="_blank" rel="ugc" href="https://github.com/hybskgks28275/ComfyUI-hybs-nodes">https://github.com/hybskgks28275/ComfyUI-hybs-nodes</a><a target="_blank" rel="ugc" href="https://github.com/hybskgks28275/ComfyUI-Anima-NAG">https://github.com/hybskgks28275/ComfyUI-Anima-NAG</a><a target="_blank" rel="ugc" href="https://github.com/hybskgks28275/ComfyUI-LoRA-Merge-EXPERIMENTAL">https://github.com/hybskgks28275/ComfyUI-LoRA-Merge-EXPERIMENTAL</a><ul><li>Build Tools for Visual Studio</li></ul>If you enable torch.compile, please install triton-windows.Also, install Build Tools for Visual Studio and make sure that cl.exe is added to your system PATH.

Workflow for MOSS Image/Anima/SDXL (NoobAI XL/Illustrious-XL)

This workflow is designed to apply a LoRA only when a regular expression match is detected.It works even more effectively when combined with comfyui-dynamicprompts.Please place a .toml file in advance in the custom_nodes\ComfyUI-hybs-nodes\config directory.This file should describe the conditions under which the LoRA is applied.Please refer to the way conditions are written in the following repository:<a target="_blank" rel="ugc" href="https://github.com/hybskgks28275/ComfyUI-hybs-nodes">https://github.com/hybskgks28275/ComfyUI-hybs-nodes</a>

Files with SHA256: 18978c0ddb39a62542ec1bec524112cd54eb2cada3babe6990464cdafdccbb1c