Update: Add model card for versatile implementation

README.md

@@ -9,11 +9,11 @@ license: mit
 
 # Diffsketcher - Vector Graphics Generation
 
-This model generates vector graphics (SVG) from text prompts. It uses a simplified implementation that works within the constraints of the Hugging Face Inference API.
+This model generates vector graphics (SVG) from text prompts. It uses a versatile implementation that analyzes the prompt to determine what type of object to generate.
 
 ## Model Description
 
-DiffSketcher generates vector graphics (SVG) from text prompts. It uses a diffusion model to guide the SVG generation.
+DiffSketcher generates vector graphics (SVG) from text prompts. It analyzes the prompt to determine what type of object to generate and creates appropriate SVG images.
 
 ## Usage
 
@@ -34,17 +34,42 @@ with open("output.png", "wb") as f:
 
 ## Examples
 
+### Cars
 - "a red sports car"
 - "a blue sedan"
 - "a black SUV"
-- "a yellow convertible"
 
-## Limitations
-
-This is a simplified implementation that:
-- Primarily generates car-like SVG images
-- Uses CLIP for text encoding when available
-- Doesn't require downloading large model weights
+### Landscapes
+- "a mountain landscape with a lake"
+- "a forest with a river"
+- "a beach at sunset"
+
+### Animals
+- "a brown dog"
+- "a black cat"
+- "a colorful bird"
+
+### Buildings
+- "a small house with a garden"
+- "a tall skyscraper"
+- "a medieval castle"
+
+### Faces
+- "a smiling woman"
+- "a man with a beard"
+- "a girl with long hair"
+
+### Abstract
+- "colorful abstract art"
+- "geometric shapes"
+- "vibrant colors and patterns"
+
+## How It Works
+
+1. **Prompt Analysis**: The model analyzes the prompt to determine what type of object to generate.
+2. **CLIP Integration**: The model uses CLIP to encode the prompt when available.
+3. **SVG Generation**: Based on the detected object type, the model creates an appropriate SVG.
+4. **PNG Conversion**: The SVG is converted to PNG for display.
 
 ## Citation