Update app.py

app.py

@@ -1,26 +1,23 @@
 import gradio as gr
+import spaces  # Required for ZeroGPU
 from transformers import pipeline
 from duckduckgo_search import DDGS
 from datetime import datetime
-import asyncio
 
-# Initialize a lightweight text generation model (distilgpt2 for speed)
-generator = pipeline("text-generation", model="distilgpt2", device=0 if gr.cuda.is_available() else -1)
+# Initialize a lightweight text generation model on CPU (moves to GPU when decorated)
+generator = pipeline("text-generation", model="distilgpt2", device=-1)  # -1 ensures CPU by default
 
-# Web search function using DuckDuckGo
-async def get_web_results(query: str, max_results: int = 5) -> list:
-    """Fetch web results asynchronously for deep research."""
+# Web search function (CPU-based)
+def get_web_results(query: str, max_results: int = 3) -> list:
+    """Fetch web results synchronously for Zero GPU compatibility."""
     try:
         with DDGS() as ddgs:
-            results = await asyncio.to_thread(lambda: list(ddgs.text(query, max_results=max_results)))
-            return [
-                {"title": r.get("title", "No Title"), "snippet": r["body"], "url": r["href"]}
-                for r in results
-            ]
+            results = list(ddgs.text(query, max_results=max_results))
+            return [{"title": r.get("title", "No Title"), "snippet": r["body"], "url": r["href"]} for r in results]
     except Exception as e:
         return [{"title": "Error", "snippet": f"Failed to fetch results: {str(e)}", "url": "#"}]
 
-# Format prompt for the AI model
+# Format prompt for the AI model (CPU-based)
 def format_prompt(query: str, web_results: list) -> str:
     """Create a concise prompt with web context."""
     current_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
@@ -29,70 +26,50 @@ def format_prompt(query: str, web_results: list) -> str:
 Query: {query}
 Web Context:
 {context}
-Provide a detailed, well-structured answer in markdown format with citations [1], [2], etc."""
+Provide a concise answer in markdown format with citations [1], [2], etc."""
 
-# Generate answer using the AI model
+# GPU-decorated answer generation
+@spaces.GPU(duration=120)  # Allow up to 120 seconds of GPU time
 def generate_answer(prompt: str) -> str:
-    """Generate a detailed research answer."""
-    response = generator(prompt, max_length=300, num_return_sequences=1, truncation=True)[0]["generated_text"]
-    # Extract the answer after the prompt
-    answer_start = response.find("Provide a detailed") + len("Provide a detailed, well-structured answer in markdown format with citations [1], [2], etc.")
-    return response[answer_start:].strip()
+    """Generate a concise research answer using GPU."""
+    response = generator(prompt, max_length=200, num_return_sequences=1, truncation=True)[0]["generated_text"]
+    answer_start = response.find("Provide a concise") + len("Provide a concise answer in markdown format with citations [1], [2], etc.")
+    return response[answer_start:].strip() if answer_start > -1 else "No detailed answer generated."
 
-# Format sources for display
+# Format sources for display (CPU-based)
 def format_sources(web_results: list) -> str:
-    """Create an HTML list of sources."""
+    """Create a simple HTML list of sources."""
     if not web_results:
         return "<div>No sources available</div>"
-    
     sources_html = "<div class='sources-list'>"
     for i, res in enumerate(web_results, 1):
         sources_html += f"""
         <div class='source-item'>
             <span class='source-number'>[{i}]</span>
-            <a href='{res['url']}' target='_blank'>{res['title']}</a>: {res['snippet'][:150]}...
+            <a href='{res['url']}' target='_blank'>{res['title']}</a>: {res['snippet'][:100]}...
         </div>
         """
     sources_html += "</div>"
     return sources_html
 
 # Main processing function
-async def process_deep_research(query: str, history: list):
-    """Handle the deep research process with progressive updates."""
+def process_deep_research(query: str, history: list):
+    """Handle the deep research process."""
     if not history:
         history = []
 
-    # Step 1: Initial loading state
-    yield {
-        "answer": "*Searching the web...*",
-        "sources": "<div>Fetching sources...</div>",
-        "history": history + [[query, "*Searching...*"]]
-    }
-
-    # Step 2: Fetch web results
-    web_results = await get_web_results(query)
+    # Fetch web results (CPU)
+    web_results = get_web_results(query)
     sources_html = format_sources(web_results)
 
-    # Step 3: Update with web search completed
-    yield {
-        "answer": "*Analyzing results...*",
-        "sources": sources_html,
-        "history": history + [[query, "*Analyzing...*"]]
-    }
-
-    # Step 4: Generate detailed answer
+    # Generate answer (GPU via @spaces.GPU)
     prompt = format_prompt(query, web_results)
     answer = generate_answer(prompt)
     final_history = history + [[query, answer]]
 
-    # Step 5: Final result
-    yield {
-        "answer": answer,
-        "sources": sources_html,
-        "history": final_history
-    }
+    return answer, sources_html, final_history
 
-# Custom CSS for a cool, modern UI
+# Custom CSS for a cool, lightweight UI
 css = """
 body {
     font-family: 'Arial', sans-serif;
@@ -100,31 +77,24 @@ body {
     color: #ffffff;
 }
 .gradio-container {
-    max-width: 1000px;
+    max-width: 900px;
     margin: 0 auto;
-    padding: 20px;
+    padding: 15px;
 }
 .header {
     text-align: center;
-    padding: 20px;
+    padding: 15px;
     background: linear-gradient(135deg, #2c3e50, #3498db);
-    border-radius: 10px;
-    margin-bottom: 20px;
-}
-.header h1 {
-    font-size: 2.5em;
-    margin: 0;
-    color: #ffffff;
-}
-.header p {
-    color: #bdc3c7;
-    font-size: 1.1em;
+    border-radius: 8px;
+    margin-bottom: 15px;
 }
+.header h1 { font-size: 2em; margin: 0; color: #ffffff; }
+.header p { color: #bdc3c7; font-size: 1em; }
 .search-box {
     background: #2c2c2c;
-    padding: 15px;
-    border-radius: 10px;
-    box-shadow: 0 4px 10px rgba(0, 0, 0, 0.3);
+    padding: 10px;
+    border-radius: 8px;
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.2);
 }
 .search-box input {
     background: #3a3a3a !important;
@@ -136,68 +106,50 @@ body {
     background: #3498db !important;
     border: none !important;
     border-radius: 5px !important;
-    transition: background 0.3s;
-}
-.search-box button:hover {
-    background: #2980b9 !important;
 }
 .results-container {
-    margin-top: 20px;
+    margin-top: 15px;
     display: flex;
-    gap: 20px;
+    gap: 15px;
 }
 .answer-box {
     flex: 2;
     background: #2c2c2c;
-    padding: 20px;
-    border-radius: 10px;
-    box-shadow: 0 4px 10px rgba(0, 0, 0, 0.3);
-}
-.answer-box .markdown {
-    color: #ecf0f1;
-    line-height: 1.6;
+    padding: 15px;
+    border-radius: 8px;
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.2);
 }
+.answer-box .markdown { color: #ecf0f1; line-height: 1.5; }
 .sources-list {
     flex: 1;
     background: #2c2c2c;
-    padding: 15px;
-    border-radius: 10px;
-    box-shadow: 0 4px 10px rgba(0, 0, 0, 0.3);
-}
-.source-item {
-    margin-bottom: 10px;
-}
-.source-number {
-    color: #3498db;
-    font-weight: bold;
-    margin-right: 5px;
-}
-.source-item a {
-    color: #3498db;
-    text-decoration: none;
-}
-.source-item a:hover {
-    text-decoration: underline;
-}
+    padding: 10px;
+    border-radius: 8px;
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.2);
+}
+.source-item { margin-bottom: 8px; }
+.source-number { color: #3498db; font-weight: bold; margin-right: 5px; }
+.source-item a { color: #3498db; text-decoration: none; }
+.source-item a:hover { text-decoration: underline; }
 .history-box {
-    margin-top: 20px;
+    margin-top: 15px;
     background: #2c2c2c;
-    padding: 15px;
-    border-radius: 10px;
-    max-height: 300px;
+    padding: 10px;
+    border-radius: 8px;
+    max-height: 250px;
     overflow-y: auto;
-    box-shadow: 0 4px 10px rgba(0, 0, 0, 0.3);
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.2);
 }
 """
 
-# Gradio app setup with Blocks for better control
-with gr.Blocks(title="Deep Research Engine", css=css) as demo:
+# Gradio app setup with Blocks
+with gr.Blocks(title="Deep Research Engine - ZeroGPU", css=css) as demo:
     history_state = gr.State([])
 
     # Header
     with gr.Column(elem_classes="header"):
         gr.Markdown("# Deep Research Engine")
-        gr.Markdown("Your gateway to in-depth answers with real-time web insights.")
+        gr.Markdown("Fast, in-depth answers powered by web insights (ZeroGPU).")
 
     # Search input and button
     with gr.Row(elem_classes="search-box"):
@@ -216,15 +168,15 @@ with gr.Blocks(title="Deep Research Engine", css=css) as demo:
         history_display = gr.Chatbot(label="History", elem_classes="history-box")
 
     # Event handling
-    async def handle_search(query, history):
-        async for step in process_deep_research(query, history):
-            yield step["answer"], step["sources"], step["history"]
+    def handle_search(query, history):
+        answer, sources, new_history = process_deep_research(query, history)
+        return answer, sources, new_history
 
     search_btn.click(
         fn=handle_search,
         inputs=[search_input, history_state],
         outputs=[answer_output, sources_output, history_display],
-        _js="() => [document.querySelector('.search-box input').value, null]"  # Ensure history is managed
+        _js="() => [document.querySelector('.search-box input').value, null]"
     ).then(
         fn=lambda x: x,
         inputs=[history_display],