Update README.md

README.md

@@ -4,6 +4,11 @@ license: mit
 
 # 🚀 ComfyUI Workflows by wizdroid
 
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![ComfyUI](https://img.shields.io/badge/ComfyUI-Workflows-blue)](https://github.com/comfyanonymous/ComfyUI)
+[![Flux](https://img.shields.io/badge/Flux-Accelerated-orange)](https://blackforestlabs.ai/)
+[![Nunchaku](https://img.shields.io/badge/Nunchaku-Fast%20Inference-purple)](https://github.com/nunchaku-tech)
+
 **High-quality, production-ready ComfyUI workflows** focused on **character consistency**, fast generation, and automated dataset creation for LoRA training.
 
 These workflows power consistent character generation, multi-reference image-to-image pipelines, lightning-fast turbo inference, state-of-the-art upscaling, and intelligent dataset preparation.
@@ -26,81 +31,129 @@ These workflows power consistent character generation, multi-reference image-to-
 
 ## 📁 Workflows
 
-| Workflow                  | Type          | Key Features                                      | Best For |
-|---------------------------|---------------|---------------------------------------------------|----------|
-| `flux2-klein-t2i.json`    | Text-to-Image | Nunchaku Flux2-Klein 9B, advanced Flux sampler, LoRA support | High-quality prompt-driven generation |
-| `flux2-klein-1i2i.json`   | 1-Image Ref   | Single reference image + Reference Latent         | Strong character consistency from one photo |
-| `flux2-klein-2i2i.json`   | 2-Image Ref   | Dual reference images                             | Even stronger identity with two angles/expressions |
-| `flux2-klein-3i2i.json`   | 3-Image Ref   | Triple reference images + advanced conditioning   | **Maximum character fidelity** (recommended for LoRA training) |
-| `qwen-t2i.json`           | Text-to-Image | Qwen-based, flexible resolution, LoRA             | Alternative aesthetic / artistic control |
-| `qwen-aio-t2i.json`       | Text-to-Image | Qwen All-in-One variant                           | Quick Qwen generations |
-| `qwen-aio-1i2i.json`      | 1-Image Ref   | Qwen image reference                              | Qwen-powered character consistency |
-| `qwen-aio-2i2i.json`      | 2-Image Ref   | Dual reference                                    | Multi-ref Qwen |
-| `qwen-aio-3i2i.json`      | 3-Image Ref   | Triple reference                                  | Maximum Qwen consistency |
-| `z-image-turbo-t2i.json`  | Fast T2I      | Z-Image-Turbo (very few steps), Nunchaku          | Lightning-fast prototyping & iteration |
-| `seedvr2-i2i.json`        | Upscaler      | SeedVR2 DiT + VAE upscaler, tiling, color correction | Best-in-class 2×–4× upscaling & detail recovery |
-| `dataset-generator.json`  | Dataset Tool  | WizdroidLoRADataset + Ollama vision (Moondream)   | Automated LoRA training dataset creation with captions |
+| Workflow                  | Type          | Size    | Key Features                                      | Best For |
+|---------------------------|---------------|---------|---------------------------------------------------|----------|
+| `flux2-klein-t2i.json`    | Text-to-Image | ~23 MB  | Nunchaku Flux2-Klein 9B, advanced Flux sampler, LoRA support | High-quality prompt-driven generation |
+| `flux2-klein-1i2i.json`   | 1-Image Ref   | ~29 MB  | Single reference image + Reference Latent         | Strong character consistency from one photo |
+| `flux2-klein-2i2i.json`   | 2-Image Ref   | ~35 MB  | Dual reference images                             | Even stronger identity with two angles/expressions |
+| `flux2-klein-3i2i.json`   | 3-Image Ref   | ~41 MB  | Triple reference images + advanced conditioning   | Maximum character fidelity (recommended for LoRA training) |
+| `qwen-t2i.json`           | Text-to-Image | ~14 MB  | Qwen-based, flexible resolution, LoRA             | Alternative aesthetic / artistic control |
+| `qwen-aio-t2i.json`       | Text-to-Image | ~12 MB  | Qwen All-in-One variant                           | Quick Qwen generations |
+| `qwen-aio-1i2i.json`      | 1-Image Ref   | ~14 MB  | Qwen image reference                              | Qwen-powered character consistency |
+| `qwen-aio-2i2i.json`      | 2-Image Ref   | ~15 MB  | Dual reference                                    | Multi-ref Qwen |
+| `qwen-aio-3i2i.json`      | 3-Image Ref   | ~16 MB  | Triple reference                                  | Maximum Qwen consistency |
+| `z-image-turbo-t2i.json`  | Fast T2I      | ~18 MB  | Z-Image-Turbo (very few steps), Nunchaku          | Lightning-fast prototyping & iteration |
+| `seedvr2-i2i.json`        | Upscaler      | ~13 MB  | SeedVR2 DiT + VAE upscaler, tiling, color correction | Best-in-class 2×–4× upscaling & detail recovery |
+| `dataset-generator.json`  | Dataset Tool  | ~3 MB   | WizdroidLoRADataset + Ollama vision (Moondream)   | Automated LoRA training dataset creation with captions |
 
 ---
 
 ## 🛠️ Installation & Setup
 
 ### 1. ComfyUI
-Recent [ComfyUI](https://github.com/comfyanonymous/ComfyUI) installation.
+Make sure you have a recent [ComfyUI](https://github.com/comfyanonymous/ComfyUI) installation.
 
 ### 2. Required Custom Nodes
-Install via **ComfyUI Manager** (recommended):
+Install these via **ComfyUI Manager** (recommended) or git clone into `custom_nodes/`:
 
-- **ComfyUI-nunchaku** — Accelerated Flux inference
-- **ComfyUI-SeedVR2_VideoUpscaler** (ainvfx)
-- **Wizdroid Character Dataset Node** (WizdroidLoRADataset and related nodes)
+- **ComfyUI-nunchaku** — For accelerated Flux inference
+- **ComfyUI-SeedVR2_VideoUpscaler** (ainvfx) — For the SeedVR2 upscaler workflow
+- **Wizdroid Character Dataset Node** — The custom `WizdroidLoRADataset` node (included in this repo's ecosystem)
 
-Restart ComfyUI after installing.
+Restart ComfyUI after installing custom nodes.
 
 ### 3. Models
-Place in standard ComfyUI folders:
+Place models in the standard ComfyUI folders (or update paths inside workflows):
+
+**Flux2-Klein (main quality engine):**
+- `flux/2/flux2-klein-9B.safetensors`
+- `flux/2/vae.safetensors`
+
+**Qwen:**
+- `qwen/qwen-image-edit-sfw.safetensors`
+- `qwen/qwen3-8B.safetensors` (or 4B variant)
+- `qwen/qwen3-4B.safetensors`
+
+**Z-Image Turbo:**
+- `z-image/z-image-turbo.safetensors`
+- `z-image/vae.safetensors`
+- `z-image/turbo/4nup4m4.safetensors` (example turbo LoRA)
 
-- **Flux2-Klein**: `flux/2/flux2-klein-9B.safetensors` + VAE
-- **Qwen** models (as referenced in workflows)
-- **Z-Image-Turbo** + turbo LoRAs
-- **SeedVR2**: `seedvr2_ema_7b_sharp_fp16.safetensors` + VAE
+**SeedVR2:**
+- `seedvr2_ema_7b_sharp_fp16.safetensors` (DiT)
+- `ema_vae_fp16.safetensors` (VAE)
+
+**CLIP / Other:**
+- Various Qwen CLIP models as referenced in the workflows
+
+### 4. Ollama (for Dataset Generator)
+The `dataset-generator.json` workflow uses a local vision model via Ollama:
 
-### 4. Ollama (Dataset Generator)
 ```bash
+# Install Ollama and pull a good vision model (Moondream recommended in the workflow)
 ollama pull moondream:latest
-ollama serve
 ```
 
+Start Ollama (`ollama serve`) before running the dataset workflow. You can change the model/URL inside the node.
+
 ---
 
 ## 🚀 How to Use
 
-1. Drag & drop any `.json` workflow onto ComfyUI.
-2. Use the clean proxy widgets for prompt, seed, steps, LoRA toggle, and reference images.
-3. For **character consistency**, load prompts from **[PROMPTS.md](https://huggingface.co/wizdroid/comfyui-workflows/blob/main/PROMPTS.md)** into the 1i2i/2i2i/3i2i workflows.
+1. **Load a workflow**: Drag & drop any `.json` file directly onto your ComfyUI canvas, or use **Load** → select the file.
+2. **Queue Prompt** as usual.
+3. Most workflows expose clean proxy widgets on the right for:
+   - Prompt / negative prompt
+   - Seed, steps, CFG, sampler
+   - Width / Height
+   - LoRA toggle + strength
+   - Reference images (for i2i workflows)
+
+### Character Consistency Workflow Recommendations
 
-**Recommended for LoRA training data:** `flux2-klein-3i2i.json` + 2–3 clean source references + the 40 reference prompts.
+- Use **flux2-klein-3i2i.json** (or 2i2i) with 2–3 high-quality, varied reference photos of your character.
+- Generate using the same character across many angles/expressions using the reference prompts collection.
+- Feed the resulting images into `dataset-generator.json` to auto-caption and structure a ready-to-train dataset (with your character trigger tag).
 
-Then run `dataset-generator.json` to auto-caption and structure everything with your trigger word.
+---
+
+## 💡 Tips for Best Results
+
+- **Reference Images**: Clean, well-lit, minimal makeup/jewelry/background. Square 1:1 references work great.
+- **Prompting**: Start simple ("Photo of a woman", "beautiful detailed face") and let the reference images do the heavy lifting for identity.
+- **LoRAs**: The LoRA switch is disabled by default in many workflows. Enable it and point to character or style LoRAs.
+- **Speed vs Quality**: Use `z-image-turbo-t2i.json` for rapid iteration, then switch to Flux2-Klein for final outputs.
+- **Upscaling**: Always run important generations through `seedvr2-i2i.json` for that extra polish.
+- **Dataset Quality**: Use the dataset generator + good reference images + Ollama captions, then manually review the `validation_report`.
 
 ---
 
-## 💡 Full Guide: Mastering Consistent Characters
+## � Full Guide: Mastering Consistent Characters
+
+For the **complete step-by-step tutorial** (problem → solution → exact pipeline using these workflows + PROMPTS.md + the WizdroidLoRADataset node → pro tips + visual examples), see the dedicated CivitAI article:
 
-For the complete step-by-step tutorial (problem → solution → pipeline with these workflows + PROMPTS.md + WizdroidLoRADataset → pro tips), see the dedicated CivitAI article:
+**→ [Mastering Consistent Characters for LoRA Training](https://civitai.com/articles/31442/mastering-consistent-characters-for-lora-training)**
 
-**→ [Mastering Consistent Characters for LoRA Training](https://civitai.com/articles/...)**  
-*(Replace with the real link after publishing the article on CivitAI)*
+The article covers:
+- Why single-reference methods fail for modern models (Flux, SD3, Aurora, etc.)
+- How Reference Latent injection + clean square 1:1 prompts solve identity drift
+- Generating 40–100+ varied training images using the included PROMPTS.md
+- Using the WizdroidLoRADataset node for perfect auto-captions, validation reports, and trigger word injection
+- Training recommendations, best practices, and real-world results
 
-It covers why single-ref fails, how Reference Latent + the 40 square prompts solve identity drift, generating 40–100+ training images, auto-captioning with the custom node, and training best practices.
+This README is the technical reference. The CivitAI guide is the teaching + visual companion. Use both.
 
-This short README is the HF model card version. The full rich version is in the local `README.md`.
+---
+
+## �📜 License
+
+MIT License — feel free to use, modify, and share these workflows.
 
 ---
 
-## 📜 License
+## 🤝 Contributing
 
-MIT — use, modify, and share freely.
+Pull requests and workflow improvements are welcome! If you create amazing results with these, tag **@wizdroid** — I'd love to see what you build.
 
 ---
 
@@ -112,4 +165,7 @@ MIT — use, modify, and share freely.
 
 **Additional Resources**
 - [PROMPTS.md — 40 Consistent Character Reference Prompts](https://huggingface.co/wizdroid/comfyui-workflows/blob/main/PROMPTS.md)
-- Full workflows and updates in this repository
+- [CivitAI Guide: Mastering Consistent Characters for LoRA Training](https://civitai.com/articles/...) *(update this link after posting the article)*
+- Full workflows and updates in this repository31442/mastering-consistent-characters-for-lora-training)
+
+**For HF users:** The `HF_README_UPDATED.md` file in this folder is a clean, concise version you can copy directly into the repo root as `README.md` if you prefer a shorter model card. The current README is the rich version recommended for the repo.