Update README.md

README.md

@@ -1,10 +1,11 @@
+
 ---
 title: GitHub Graphics Generator
 emoji: 🎨
 colorFrom: blue
 colorTo: purple
 sdk: gradio
-sdk_version: 6.2.0
+sdk_version: 4.44.1
 app_file: app.py
 pinned: false
 license: mit
@@ -51,4 +52,62 @@ The result is **consistent, readable, and reusable graphics**.
 ## How to Use
 
 1. Paste a GitHub repository URL  
-   Example:
+   Example:
+
+https://github.com/owner/repo
+
+2. Choose:
+- Style (Modern, Minimal, Professional)
+- Color scheme
+- Output size
+
+3. Click **Generate**
+
+4. Download the generated PNG
+
+---
+
+## Supported Models
+
+The following Hugging Face Inference API models are supported:
+
+- `stabilityai/stable-diffusion-xl-base-1.0`
+- `runwayml/stable-diffusion-v1-5`
+
+> Note: Diffusers-only models (e.g. SDXL-Lightning) are intentionally not used to ensure compatibility with Hugging Face Spaces.
+
+---
+
+## Technical Overview
+
+- **Frontend**: Gradio (Blocks API)
+- **Backend**: Python
+- **AI Inference**: Hugging Face `InferenceClient`
+- **Layout & Text Rendering**: Pillow (PIL)
+- **Data Source**: GitHub REST API
+
+---
+
+## Environment Variables
+
+Optional but recommended:
+
+```env
+HUGGINGFACE_TOKEN=hf_xxxxxxxxxxxxxxxxx
+
+This enables:
+	•	Higher rate limits
+	•	Faster inference
+	•	Access to gated models (if applicable)
+
+⸻
+
+Known Limitations
+	•	AI does not render text or logos
+	•	GitHub API rate limits apply without a token
+	•	Image generation speed depends on model availability
+
+⸻
+
+License
+