⚡ Illustrious Workflow as of version 18f
🛠️ Purpose & Design Philosophy
This workflow is a high-fidelity environment built for Illustrious XL. It prioritizes stability and professional texture over generation speed. It follows an "all-in-one" philosophy: configure your prompts, hit queue, and let the workflow handle the multi-stage refinement from start to finish.
Not for Speed: This is a heavy-duty refinement tool. If you want 2-second previews, use a basic SDXL workflow.
Personal Use: Built for my specific production needs. It is shared as-is for those who want a "set-and-forget" pipeline for Illustrious. Adjust the settings and models to fit your needs.
All-in-One Logic: The workflow handles generation, detailing, and upscaling in one continuous pass.
If you're needing all the bells and whistles: Try checking out ComfyUI Image Workflows by Legendaer here on CivitAI.
🚀 Key Features & 2026 Logic
Global Model Patching (RescaleCFG): Includes a pre-configured RescaleCFG patch (Multiplier: 0.7) applied globally. This acts as "HDR Insurance," preventing the "deep-fried" or over-saturated look common in high-CFG Illustrious runs.
Detail Daemon Sampler: Integrated to enhance structural depth. In this version, it is tuned to start at 0.4 to preserve the core Illustrious character proportions while sharpening hair and eye details.
Hybrid Upscale Strategy: * Group Bypass Switch: Easily toggle between a RTX VSR path for flat anime styles and an Upscale-Model path for 2.5D/highly detailed renders.
Ultimate SD Upscale: Re-draws the upscaled canvas at a 0.2 to 0.35 denoise to lock in fine textures.
Power LoRA Loader: Manage multiple Illustrious-specific LoRAs without messy wiring.
Triple Detailer Groups: 3-stage targeted refinement for faces, hands, and clothing using standard detection models.
CivitAI Meta-Sync: Images are saved with full metadata (Model, LoRAs, Sampler info) for automatic site parsing.
⚠️ Disclaimer & Compatibility
Install at Your Own Risk: Custom nodes can break your environment. I am not responsible for troubleshooting your specific installation.
ComfyUI Portable: Built and tested on the Portable version. Desktop app users may face additional hurdles.
The "Your Version" Factor: Your node versions and environment are 99.9% likely to differ from mine.
Nodes 2.0: I do not recommend using Nodes 2.0. It creates unpredictable UI behavior; I will not provide support for issues involving this feature.
🤝 Support & Boundaries
No DMs: DMs are disabled due to repeat spam. Please check the Discussions tab below; most questions have already been answered.
Modifications: You are free to hack this workflow apart. However, you are responsible for fixing it if it breaks.
Custom Requests: I do not make private workflows. If you need a custom solution, post a Bounty on CivitAI. There are many talented creators ready to help you for a fee.
🛠️ Other Related Info
ComfyUI version: v0.19.3
ComfyUI frontend: v.1.43.1
Custom node links for manual install:
Description
Illustrious Workflow v18e — Release Notes
This update focuses on refining noise handling, restoring high-utility scheduling options, and providing more flexibility in your upscaling pipeline. Based on extensive testing, several "quality of life" toggles have been added to help you dial in the perfect balance between speed and high-fidelity output.
Core Changes & Optimizations
Removal of Epsilon Scaling Following comparative testing across multiple seeds and configurations (testing Epsilon Scaling vs. CFGZeroStar), it was determined that CFGZeroStar alone provides the cleanest, most stable results. Epsilon Scaling has been removed to reduce overhead and simplify the generation path.
Restored: ModelSamplingDiscrete Re-added specifically to allow for granular control over ZSNR. You can now toggle Zero Terminal SNR on or off depending on your specific contrast and detail requirements. Note: this does not work with all models, samplers, or schedulers.
Restored: Laplace Scheduler Now housed within its own dedicated subgraph. It includes a toggle switch for easy integration or bypassing without disrupting the rest of your scheduling logic.
Added: Sigmas Rescale (from RES4LYF) Integrated to improve initial noise handling. This provides a more robust alternative to standard denoising on the base image, leading to better structural integrity in the early stages of generation. Note: this does not work with all samplers/schedulers.
Updated Upscale Pipeline
Pre-USDU Upscale Flexibility You no longer have to choose one method during setup. Both RTX VSR and Standard Model-Based Upscaling are now integrated into the same workflow.
Toggle Switch: A new hardware-level switch allows you to swap between these two methods instantly.
The Trade-off: Use RTX VSR when speed is your primary goal; use Model Upscaling when you are prioritizing maximum texture quality and fidelity.
Please note: if you do not have an RTX video card you can just delete the RTX nodes from the workflow. You might have to unpack the sub-graph in the Pre-USDU group and delete the switches if an error occurs when deleting the RTX nodes.
Workflow Navigation & Usability
On-Canvas Documentation: New Markdown Notes have been placed directly on the ComfyUI canvas. these provide quick-reference instructions on how the newly added and restored nodes function.
Default State: To keep the workflow "lean" and fast for standard use, all "extra" settings are bypassed by default.
Fast Access: Use the Fast Groups Bypasser nodes (located on the far left of the workflow) to quickly enable or disable these advanced features.