Update: Add model card for simplified implementation

README.md

@@ -1,22 +1,19 @@
 ---
-license: mit
-tags:
-- text-to-image
-- svg
-- vector-graphics
 pipeline_tag: text-to-image
-widget:
-- text: "a beautiful mountain landscape"
-  example_title: "Mountain Landscape"
-- text: "a cute cat sitting on a windowsill"
-  example_title: "Cat on Windowsill"
-- text: "a futuristic city skyline at sunset"
-  example_title: "Futuristic City"
+tags:
+  - text-to-image
+  - svg
+  - vector-graphics
+license: mit
 ---
 
-# Vector Graphics Model
+# Diffsketcher - Vector Graphics Generation
 
-This model generates vector graphics (SVG) from text prompts.
+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.
+
+## Model Description
+
+DiffSketcher generates vector graphics (SVG) from text prompts. It uses a diffusion model to guide the SVG generation.
 
 ## Usage
 
@@ -26,25 +23,40 @@ import requests
 API_URL = "https://api-inference.huggingface.co/models/jree423/diffsketcher"
 headers = {"Authorization": "Bearer YOUR_API_TOKEN"}
 
-def query(payload):
-    response = requests.post(API_URL, headers=headers, json=payload)
-    return response.json()
-
-# Text-to-image input
-output = query({
-    "inputs": "a beautiful mountain landscape",
-})
+def query(prompt):
+    response = requests.post(API_URL, headers=headers, json={"inputs": prompt})
+    return response.content  # Returns the image directly
 
-# The response contains:
-# - svg: The SVG content as a string
-# - svg_base64: The SVG content encoded as base64
-# - png_base64: A PNG version of the SVG encoded as base64
+# Generate an image
+with open("output.png", "wb") as f:
+    f.write(query("a red sports car"))
 ```
 
-## Model Details
+## Examples
 
-This model generates vector graphics (SVG) from text prompts. It produces clean, scalable vector graphics that can be used in various applications.
+- "a red sports car"
+- "a blue sedan"
+- "a black SUV"
+- "a yellow convertible"
 
 ## Limitations
 
-This is a placeholder implementation that returns simple SVG graphics. The full model implementation will be added in the future.
+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
+
+## Citation
+
+```
+@article{xing2023diffsketcher,
+  title={{DiffSketcher}: Text Guided Vector Sketch Synthesis through Latent Diffusion Models},
+  author={Xing, XiMing and Zhan, Chuang and Xu, Yinghao and Dong, Yue and Yu, Yingqing and Li, Chongyang and Liu, Yongyi and Ma, Chongxuan and Tao, Dacheng},
+  journal={arXiv preprint arXiv:2306.14685},
+  year={2023}
+}
+```
+
+## License
+
+This model is licensed under the MIT License.