File size: 8,917 Bytes
a4f2dc1 9b49d4e a4f2dc1 9b49d4e a4f2dc1 9b49d4e a4f2dc1 9b49d4e a4f2dc1 9b49d4e a4f2dc1 4797a38 a4f2dc1 9b49d4e a4f2dc1 3614ae1 111eecc 3614ae1 a4f2dc1 3614ae1 a4f2dc1 3614ae1 a4f2dc1 3614ae1 a4f2dc1 4797a38 a4f2dc1 9b49d4e a4f2dc1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 |
---
license: apache-2.0
language:
- en
- zh
base_model:
- Tongyi-MAI/Z-Image-Turbo
pipeline_tag: text-to-image
tags:
- text-to-image
- image-generation
- diffusion
- comfyui
- photorealistic
- bilingual
- chinese
- english
- 8-step
- fast-generation
---
# ๐ Z-Image-Turbo-AIO | 8-Step Photorealistic Generation
<div align="center">
**Ultra-Fast โข Bilingual Text Rendering โข All-in-One โข FP8 & BF16**
[](https://opensource.org/licenses/Apache-2.0)
[](https://github.com/comfyanonymous/ComfyUI)
</div>
## โจ What is Z-Image-Turbo-AIO?
Z-Image-Turbo-AIO is an **All-in-One repackage** of Alibaba Tongyi Lab's 6B parameter photorealistic image generator, optimized for lightning-fast 8-step generation. This version includes **integrated VAE and Text Encoder** for maximum convenience - just download and generate!
### Available Versions
| Version | Size | Best For |
|---------|------|----------|
| ๐ก **FP8-AIO** | ~10GB | For most users |
| ๐ **BF16-AIO** | ~20GB | Maximum quality |
## ๐ฏ Key Features
- โก **8-step generation** - 10-40 seconds per image, depends on your GPU
- ๐ฆ **All-in-One** - No separate VAE/Text Encoder downloads needed
- ๐ธ **Photorealistic** - Professional quality output
- ๐ **Bilingual** - English & Chinese text rendering
- ๐ฏ **8GB VRAM** - Works on GPUs with 8GB VRAM
- ๐ **Apache 2.0** - Open license for any use
## ๐ Which Version Should I Choose?
### ๐ก FP8-AIO (Recommended for most users)
- โ
Half the file size
- โ
Faster download
- โ
Excellent quality
- โ
Perfect for 8GB VRAM
- โ
Great for testing & everyday use
### ๐ BF16-AIO (Maximum precision)
- โ
BFloat16 full precision
- โ
Lossless quality
- โ
Great for testing & everyday use
- โ
Still works on 8GB VRAM
### ๐ผ๏ธ [CivitAI Page](https://civit.ai/models/2173571)
## ๐ฅ Quick Start (ComfyUI)
### Installation
1. Download your preferred version (FP8 or BF16)
2. Place in `ComfyUI/models/checkpoints`
3. Load with "Load Checkpoint" node
4. Generate!
---
## ๐ง ComfyUI Workflow

Download: [z-image-turbo-aio-workflow.json](z-image-turbo-aio-workflow.json)
---
### ๐ Standard Workflow v2 (Recommended)
Text-to-image generation with improved upscaler and metadata saving.

๐ฅ **Download:** [Z-Image-Turbo-AIO-workflow-v2.json](workflow/Z-Image-Turbo-AIO-workflow-v2.json)
**Features:**
- Pre-configured settings (9 steps, CFG 1.0)
- Dual samplers: `res_multistep` (sharp) or `euler_ancestral` (natural)
- Improved upscaler with denoise control
- Automatic metadata saving for easy CivitAI uploads
**Required Nodes:** `rgthree-comfy`, `comfyui_image_metadata_extension`
---
### ๐ฎ ControlNet Workflow
Precise control with reference images using ControlNet Union.

๐ฅ **Download:** [Z-Image-Turbo-AIO-workflow-controlnet.json](workflow/Z-Image-Turbo-AIO-workflow-controlnet.json)
**Features:**
- 5 control types: Canny, Depth, Pose, HED, MLSD
- Megapixel scaling (auto aspect ratio)
- ControlNet strength adjustment
**Required Nodes:** `rgthree-comfy`, `comfyui_image_metadata_extension`, `comfyui_controlnet_aux`
**Additional Download Required:**
- [Z-Image-Turbo-Fun-Controlnet-Union.safetensors](https://huggingface.co/Comfy-Org/z_image_turbo/blob/main/split_files/model_patches/z_image_turbo_fun_controlnet_union.safetensors) (~2.5GB)
- โ ๏ธ Save in: `ComfyUI/models/model_patches/` (NOT controlnet/)
- โ ๏ธ Requires ComfyUI 3.77+
---
### โ๏ธ Recommended Settings (Both Workflows)
| Parameter | Value |
|-----------|-------|
| Steps | 9 |
| CFG | 1.0 |
| Sampler | `res_multistep` (sharp) or `euler_ancestral` (natural) |
| Scheduler | `simple` (clean) or `beta` (smooth) |
| Resolution | 1920ร1088 |
๐ก **Tip:** NO negative prompts needed - the model ignores them!
## ๐ Performance
All tests on **RTX 4060 (8GB VRAM)** โข FP8 โข 1920ร1088 โข 8 steps
| Test | Generation Time |
|------|-----------------|
| Urban Interior | ~32s |
| Architecture | ~32-34s |
| Food Photography | ~32s |
| Bilingual Signage | ~32s |
---
## ๐ Test Results
All tests on **RTX 4060 (8GB VRAM)** โข FP8 โข 1920ร1088 โข 8 steps โข CFG 1.0 โข res_multistep + simple
### ๐ฌ Test 1: Urban Coffee Shop Interior

**Prompt:**
```
Modern coffee shop interior with industrial design. Exposed brick walls,
wooden beams on ceiling, pendant lights hanging above bar. Professional
espresso machine on marble counter, barista preparing latte art. Customers
sitting at wooden tables with laptops. Large windows showing city street
outside. Warm afternoon lighting, cozy atmosphere. Photorealistic style,
professional architectural photography, 8K detail.
```
โฑ๏ธ **Time:** 31.98s | ๐ฏ **Use Case:** Architectural interior photography, commercial spaces
---
### ๐ฌ Test 2: Traditional Chinese Architecture

**Prompt:**
```
Beautiful traditional Chinese temple courtyard during golden hour. Red
wooden pillars with intricate gold carvings, curved tile roofs with
upturned eaves. Stone lion statues flanking entrance. Cherry blossoms
in full bloom around courtyard. Red lanterns hanging from eaves. Soft
sunset light casting warm glow. Ancient architecture, peaceful atmosphere.
Professional travel photography, ultra-sharp detail, cinematic composition.
```
โฑ๏ธ **Time:** 33.59s | ๐ฏ **Use Case:** Travel photography, cultural heritage documentation
---
### ๐ฌ Test 3: Gourmet Food Photography

**Prompt:**
```
Professional food photography of gourmet sushi platter on black slate plate.
Assorted nigiri and maki rolls with fresh salmon, tuna, and avocado.
Garnished with pickled ginger, wasabi, and microgreens. Chopsticks placed
beside plate. Rustic wooden table surface. Soft natural window light from
side creating subtle shadows. Shallow depth of field, appetizing presentation.
Restaurant-quality styling, commercial food photography, magazine-worthy.
```
โฑ๏ธ **Time:** 32.16s | ๐ฏ **Use Case:** Food photography, restaurant menus, commercial advertising
## ๐ก Prompting Guide
### โ
Natural Language Works Best!
**Good Example:**
```
A cozy bookstore with floor-to-ceiling wooden shelves filled with
colorful books, comfortable reading nooks with cushions near large
windows, warm pendant lighting, peaceful afternoon atmosphere,
professional interior photography
```
**Bad Example:**
```
bookstore, books, chairs, window, cozy, warm light, interior
```
### ๐ Bilingual Text Rendering
**English Text:**
```
Neon sign reading "OPEN 24/7" in bright blue letters above entrance.
Modern sans-serif font, glowing effect against brick wall.
```
**Chinese Text:**
```
Traditional tea house entrance with sign reading "ๅค้ต่ถๅ" in elegant
gold Chinese calligraphy on red wooden board with ornate carved border.
```
**Both Languages:**
```
Modern cafe exterior with bilingual sign. "Morning Brew Coffee" in
white elegant script above, "ๆจๆฆๅๅก" in matching Chinese characters
below. Both glowing warmly at dusk.
```
### ๐ Prompting Tips
| Do โ
| Don't โ |
|------|---------|
| Use natural language descriptions | Use tag-style prompts (tag1, tag2) |
| Be detailed (100-300 words optimal) | Write very short prompts (<50 words) |
| Include lighting and mood | Add negative prompts (not used) |
| Describe camera angle and style | Include conflicting instructions |
| Specify materials and colors | |
## ๐ Credits & Acknowledgments
### Original Model
- **Developer:** Tongyi Lab (Alibaba Group)
- **Architecture:** Single-Stream Diffusion Transformer (6B parameters)
- **Algorithm:** Decoupled-DMD + DMDR
- **License:** Apache 2.0
### AIO Conversion
- **Created by:** [SeeSee21](https://huggingface.co/SeeSee21)
- **Format:** Integrated VAE + Text Encoder
- **Purpose:** Simplified single-file deployment
### Resources
- ๐ค [Original HuggingFace](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo)
- ๐ป [GitHub Repository](https://github.com/Tongyi-MAI/Z-Image)
- ๐จ [ComfyUI Files](https://huggingface.co/Comfy-Org/z_image_turbo)
- ๐ผ๏ธ [CivitAI Page](https://civit.ai/models/2173571)
## ๐ Version History
### v1.0 - Initial AIO Release
- FP8-AIO version (10GB)
- BF16-AIO version (20GB)
- Integrated VAE + Text Encoder
- Single-file deployment
- Based on Tongyi-MAI/Z-Image-Turbo
- Tested on RTX 4060 8GB
- Optimized for 1920ร1088
---
<div align="center">
**Download, load with "Load Checkpoint", and generate professional photos in seconds! ๐**
</div> |