Update model card with benchmarks, badges, and workflow instructions

README.md

@@ -2,45 +2,95 @@
 license: other
 license_name: qwen-research
 license_link: https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE
+language:
+- en
+- zh
 tags:
 - comfyui
 - gguf
-- diffusion-single-file
+- qwen
+- qwen2
+- qwen-image
+- qwen-image-2.1
 - text-to-image
-base_model:
-- Qwen/Qwen-Image-2.1
+- image-to-image
+- image-generation
+- quantized
+- diffusion
+pipeline_tag: text-to-image
+base_model: Qwen/Qwen-Image-2.1
+base_model_relation: quantized
 ---
 
-# Qwen-Image-2.1-GGUF
+# Qwen-Image 2.1 GGUF (ComfyUI)
 
-Direct GGUF conversion of [Qwen/Qwen-Image-2.1](https://huggingface.co/Qwen/Qwen-Image-2.1) ([Comfy-Org repackaged weights](https://huggingface.co/Comfy-Org/Qwen-Image-2.1)).
+<div align="center">
 
-These model files are optimized for use in [ComfyUI](https://github.com/comfyanonymous/ComfyUI) using the [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) custom node ().
+[![Downloads](https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fmodels%2FAlperKTS%2FQwen-Image-2.1-GGUF&query=%24.downloads&label=Downloads&color=blue&logo=huggingface)](https://huggingface.co/AlperKTS/Qwen-Image-2.1-GGUF)
+[![Likes](https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fmodels%2FAlperKTS%2FQwen-Image-2.1-GGUF&query=%24.likes&label=Likes&color=red&logo=huggingface)](https://huggingface.co/AlperKTS/Qwen-Image-2.1-GGUF)
+[![ComfyUI](https://img.shields.io/badge/ComfyUI-GGUF-green.svg)](https://github.com/city96/ComfyUI-GGUF)
+[![License](https://img.shields.io/badge/License-Qwen_Research-blue.svg)](https://huggingface.co/Qwen/Qwen-Image-2.1/blob/main/LICENSE)
 
-## Models
+</div>
 
-| File Name | Quantization | Size | Description |
+Quantized GGUF weights of the state-of-the-art **Qwen-Image 2.1** diffusion model for **ComfyUI** using [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF).
+
+These models allow running Qwen-Image 2.1 on consumer GPUs with significantly reduced VRAM footprint while preserving maximum visual quality, text-rendering capabilities, and composition fidelity.
+
+---
+
+## 📦 Quantization Versions
+
+| File | Size | VRAM Recommended | Description |
 | :--- | :--- | :--- | :--- |
-| `qwen_image_2.1_Q8_0.gguf` | Q8_0 | ~7.10 GB | 8-bit quantized model preserving 99.9% quality, with F32 1D norms and BF16 sensitive I/O projection layers |
+| **`qwen_image_2.1_Q8_0.gguf`** | ~7.1 GB | 12GB+ | Near-lossless precision. Best quality & fidelity. |
+| **`qwen_image_2.1_Q6_K.gguf`** | ~5.6 GB | 8GB - 12GB | Optimal balance of high quality and memory efficiency. |
+| **`qwen_image_2.1_Q5_K_M.gguf`** | ~4.8 GB | 8GB | Fast inference, lowest VRAM consumption, excellent quality. |
 
-## Setup & File Placement in ComfyUI
+*Note: The original BF16 diffusion model is ~14.2 GB.*
 
-Place the downloaded files in your ComfyUI directory:
+---
 
+## 🚀 How to Use in ComfyUI
+
+### 1. Install Prerequisites
+Make sure you have [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) installed in your `custom_nodes` folder:
+```bash
+cd ComfyUI/custom_nodes
+git clone https://github.com/city96/ComfyUI-GGUF
 ```
-📂 ComfyUI/
-└── 📂 models/
-    ├── 📂 diffusion_models/ (or models/unet/)
-    │   └── qwen_image_2.1_Q8_0.gguf
-    ├── 📂 text_encoders/
-    │   └── qwen3vl_8b_bf16.safetensors (or quantized variant)
-    └── 📂 vae/
-        └── qwen_image_2.1_vae_bf16.safetensors
-```
 
-## Usage in ComfyUI
+### 2. Download Required Files
+
+1. **Diffusion Model (GGUF):**
+   - Download any of the `.gguf` files from this repo (e.g. `qwen_image_2.1_Q8_0.gguf`).
+   - Place it in: `ComfyUI/models/diffusion_models/`
+
+2. **VAE:**
+   - Download `qwen_image_2.1_vae_bf16.safetensors` from [Comfy-Org/Qwen-Image-2.1](https://huggingface.co/Comfy-Org/Qwen-Image-2.1/tree/main).
+   - Place it in: `ComfyUI/models/vae/`
+
+3. **Text Encoder / CLIP:**
+   - Download `qwen_2.5_vl_7b_instruct_fp8_scaled.safetensors` (or compatible Qwen2.5-VL CLIP).
+   - Place it in: `ComfyUI/models/clip/`
+
+---
+
+## 🎨 Workflows
+
+Pre-configured workflows are included in the [`workflows/`](./workflows) folder:
+- **`Qwen_Image_2.1_GGUF_Text_to_Image.json`**: Ready-to-run ComfyUI workflow using `UnetLoaderGGUF`.
+- Drag and drop the workflow JSON directly into ComfyUI to start generating immediately.
+
+### Key Node Configuration
+- Node: **`UnetLoaderGGUF`** (from ComfyUI-GGUF)
+- `unet_name`: Select `qwen_image_2.1_Q8_0.gguf` (or `Q6_K` / `Q5_K_M`)
+- Connect to standard KSampler / ModelSamplingDiscrete nodes.
+
+---
 
-1. Install [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF).
-2. Use the **`Unet Loader (GGUF)`** node instead of the standard `UNETLoader` or `DiffusionModelLoader`.
-3. Select `qwen_image_2.1_Q8_0.gguf`.
-4. Connect the MODEL output to your KSampler / Guider pipeline.
+## 📜 Credits & License
+- Original model created by the **Qwen Team (Alibaba)**.
+- GGUF conversion & quantization by **AlperKTS**.
+- Compatible with [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) by [city96](https://github.com/city96).
+- Licensed under the original Qwen Research License.