Update README.md

README.md

@@ -23,15 +23,16 @@ Welcome to the official repository for the Z-Image（造相）project!
 
 ## 🎨 Z-Image
 
-**Z-Image** is the foundation model behind Z-Image-Turbo, designed for high-quality image generation with strong controllability, broad stylistic coverage, and support for downstream development. It serves as the primary community model in the ⚡️- Image family, while Z-Image-Turbo focuses on high-speed inference.
+**Z-Image** is the foundation model behind Z-Image-Turbo, designed for high-quality image generation with strong controllability, broad stylistic coverage, and support for downstream development.  
+It serves as the primary community model in the ⚡️- Image family, while Z-Image-Turbo focuses on high-speed inference.
 
 ### 🌟 Key Features
 
-#### 🎨 Aesthetic & Artistic Diversity
+#### 🎨 Aesthetics
 Z-Image supports a wide range of aesthetics and artistic styles, including realistic photography, anime, illustration, digital art, and stylized visuals.  
 It is suitable for creative scenarios that require rich stylistic expression rather than a single preferred aesthetic.
 
-#### 🧬 Generative Diversity
+#### 🌈 Diversity
 Z-Image emphasizes diversity across multiple generative dimensions:
 - Variations in facial identity, body pose, composition, and layout across different seeds
 - Distinct appearances for individuals in multi-person scenes
@@ -51,11 +52,11 @@ Z-Image responds strongly to negative prompts, enabling reliable suppression of
 
 | Aspect | Z-Image | Z-Image-Turbo |
 |------|------|------|
-| CFG support | Yes | No |
-| Fine-tuning | Yes | Limited |
-| Aesthetic diversity | High | Reduced |
-| Negative prompt control | Strong | None |
-| Inference speed | Slower | Faster |
+| CFG | ✅ | ❌ |
+| Steps | 50 | 8 |
+| Fintunablity | ✅ | ❌ |
+| Negative Prompting | ✅ | ❌ |
+| Diversity | High | Low |
 
 ## 🚀 Quick Start
 
@@ -78,7 +79,7 @@ HF_XET_HIGH_PERFORMANCE=1 hf download Tongyi-MAI/Z-Image
 - **Guidance scale:** 3.0 – 5.0
 - **Inference steps:** 28 – 50
 - **Negative prompts:** Strongly recommended for better control
-- **CFG normalization:** `False` for general stylism, `True` for more realism
+- **CFG normalization:** `False` for general stylism, `True` for realism
 
 ### Usage Example
 
@@ -95,7 +96,7 @@ pipe = ZImagePipeline.from_pretrained(
 pipe.to("cuda")
 
 # Generate image
-prompt = "两名年轻亚裔女性紧密站在一起，背景为朴素的灰色纹理墙面，可能是室内地毯地面。左侧女性留着长卷发，身穿藏青色毛衣，左袖有奶油色褶皱装饰，内搭白色立领衬衫，下身白色裤子；佩戴小巧金色耳钉，双臂交叉于背后。右侧女性留直肩长发，身穿奶油色卫衣，胸前印有“Tunthetables”字样，下方为“New ideas”，搭配白色裤子；佩戴银色小环耳环，双臂交叉于胸前。两人均面带微笑直视镜头。照片，自然光照明，柔和阴影，以藏青、奶油白为主的中性色调，休闲时尚摄影，中等景深，面部和上半身对焦清晰，姿态放松，表情友好，室内环境，地毯地面，纯色背景。"
+prompt = "两名年轻亚裔女性紧密站在一起，背景为朴素的灰色纹理墙面，可能是室内地毯地面。左侧女性留着长卷发，身穿藏青色毛衣，左袖有奶油色褶皱装饰，内搭白色立领衬衫，下身白色裤子；佩戴小巧金色耳钉，双臂交叉于背后。右侧女性留直肩长发，身穿奶油色卫衣，胸前印有“Tun the tables”字样，下方为“New ideas”，搭配白色裤子；佩戴银色小环耳环，双臂交叉于胸前。两人均面带微笑直视镜头。照片，自然光照明，柔和阴影，以藏青、奶油白为主的中性色调，休闲时尚摄影，中等景深，面部和上半身对焦清晰，姿态放松，表情友好，室内环境，地毯地面，纯色背景。"
 negative_prompt = "" # Optional, but would be powerful when you want to remove some unwanted content
 
 image = pipe(
@@ -103,9 +104,9 @@ image = pipe(
     negative_prompt=negative_prompt,
     height=1280,
     width=720,
-    cfg_normalization=True,
+    cfg_normalization=False,
     num_inference_steps=50,
-    guidance_scale=4.0,
+    guidance_scale=4,
     generator=torch.Generator("cuda").manual_seed(42),
 ).images[0]