README.md

---
license: creativeml-openrail-m
tags:
  - comfyui
  - model-pack
  - upscalers
  - loras
  - diffusion
  - ai-art
  - style-transfer
  - face-enhancement
---

# 🧰 ComfyUI-Models by 10xwithankit

This repository contains a curated collection of **ComfyUI-compatible models** including:

- 📈 **Upscalers** (e.g. 4x UltraSharp, RealWebPhoto)
- 🧠 **LoRAs** for style and character conditioning
- 🎨 **Skin detail and face enhancement models**
- 💠 Other diffusion-ready `.pt`, `.pth`, and `.safetensors` models

> 🚧 **Note**: This is a **personal, non-commercial hobby project**.  
> Models included here are collected from public sources like [CivitAI](https://civitai.com), [Hugging Face](https://huggingface.co), and open repositories.  
> Some models may be under third-party licenses or copyrighted by their original authors.  
> **No ownership is claimed** unless explicitly stated.

---

## 📁 Folder Structure

```plaintext
upscale_models/     → ESRGANs, Real-ESRGANs, SwinIR, etc.
loras/              → Style + character LoRAs
diffusion_models/   → Base models (e.g., SD, Flux)
controlnet/         → ControlNet modules (pose, depth, etc.)
insightface/        → Facial detection/refinement tools
...

Folder	Purpose
LLM	Language models (e.g., GGUF, transformers)
checkpoints	Stable Diffusion or custom model checkpoints
clip	CLIP models for text-image understanding
clip_vision	CLIP vision encoders for image embeddings
configs	Configuration files for models or pipelines
controlnet	ControlNet models for pose/edge/image control
diffusers	Hugging Face Diffusers-based pipelines or models
diffusion_models	Base diffusion model weights (e.g., SD, Flux)
embeddings	Text/image embeddings (e.g., IPAdapter, InsightFace)
gligen	GLIGEN components for guided generation
hypernetworks	Hypernetwork models for style or modulation
ipadapter	IPAdapter models for reference image guidance
llm_gguf	GGUF-format language models (used in llama.cpp etc.)
loras	LoRAs for styles, characters, fine-tuning
photomaker	PhotoMaker or face personalization models
style_models	Style transfer or enhancer models
text_encoders	T5, BERT, CLIPText for text tokenizing
unet	UNet backbones for SD architecture
upscale_models	ESRGAN/Real-ESRGAN/etc. upscalers
vae	VAE decoders for latent → image conversion
vae_approx	Lightweight VAE approximations



⸻

💡 Use Case

Ideal for:
	•	🎨 Artists using ComfyUI
	•	🧪 Researchers building pipelines
	•	🌍 Builders switching servers (RunPod, Vast, Colab)

Clone it anywhere:

git clone https://huggingface.co/10xwithankit/ComfyUI-Models

Download a specific model with aria2c:

aria2c -c -x 16 -s 16 "https://huggingface.co/10xwithankit/ComfyUI-Models/resolve/main/upscale_models/4xUltraSharp.pt"



⸻

🛠 How to Contribute

This is a personal model pack, but feel free to:
	•	🔁 Fork it
	•	🧩 Build on top
	•	📥 Suggest additions

⸻

📜 Disclaimer

This repository is for educational and archival purposes only.
I do not claim authorship or ownership of any third-party model unless explicitly mentioned.
If you are a model creator and wish your work to be removed or credited differently, please reach out.

⸻

Maintained by: @10xwithankit
License: CreativeML OpenRAIL-M

---