✨ Update PurifyHtml function: change model loading to use 'jinaai/ReaderLM-v2' for improved performance.

Purify.py

@@ -86,8 +86,8 @@ def PurifyHtml(Url: str) -> str: # type: ignore
 		for Line in Summary:
 			print(Line)
 
-		Tokenizer = AutoTokenizer.from_pretrained('jinaai/jina-readerLM-1.5B')
-		Model = AutoModelForCausalLM.from_pretrained('jinaai/jina-readerLM-1.5B', torch_dtype=torch.float32, device_map='cpu')
+		Tokenizer = AutoTokenizer.from_pretrained('jinaai/ReaderLM-v2')
+		Model = AutoModelForCausalLM.from_pretrained('jinaai/ReaderLM-v2', torch_dtype=torch.float32, device_map='cpu')
 
 		Prompt = f'Convert this HTML to markdown:\n\n{CleanedHtml}'
 		Inputs = Tokenizer(Prompt, return_tensors='pt', truncation=True, max_length=8192)

app.py

@@ -419,7 +419,7 @@ with gradio.Blocks(
 		
 		with gradio.TabItem('Web Scraping & Purification 🌐'):
 			with gradio.Group():
-				PurifyInput = gradio.Textbox(label='URL to Purify 🌐', placeholder='Enter URL to fetch and purify HTML', lines=1, max_lines=1)
+				PurifyInput = gradio.Textbox(label='URL to Purify 🌐', placeholder='Enter URL to fetch and purify HTML (e.g., https://huggingface.co)', lines=1, max_lines=1)
 				PurifyOutput = gradio.Text(label='Purified HTML Content 📝', interactive=False)
 				PurifyBtn = gradio.Button('Purify HTML 🧹', variant='primary')
 				PurifyBtn.click(Purify, inputs=PurifyInput, outputs=PurifyOutput)