hmmm

app.py

@@ -1,11 +1,38 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-import asyncio
+import threading
+import time
 
 # Use a smaller model
 client = InferenceClient("distilgpt2")
 
-async def respond(
+def generate_with_timeout(prompt, max_new_tokens, temperature, top_p, timeout=10):
+    result = []
+    def target():
+        try:
+            response = client.text_generation(
+                prompt,
+                max_new_tokens=max_new_tokens,
+                temperature=temperature,
+                top_p=top_p,
+                do_sample=True,
+            )
+            result.append(response)
+        except Exception as e:
+            result.append(str(e))
+
+    thread = threading.Thread(target=target)
+    thread.start()
+    thread.join(timeout)
+    
+    if thread.is_alive():
+        return "I'm sorry, but I'm having trouble generating a response right now. Could you try again?"
+    elif result:
+        return result[0]
+    else:
+        return "An error occurred while generating the response."
+
+def respond(
     message,
     history: list[tuple[str, str]],
     system_message,
@@ -19,24 +46,11 @@ async def respond(
         prompt += f"Human: {user_msg}\nAI: {bot_msg}\n"
     prompt += f"Human: {message}\nAI:"
 
-    try:
-        # Generate response with a timeout
-        response = await asyncio.wait_for(
-            client.text_generation(
-                prompt,
-                max_new_tokens=max_tokens,
-                temperature=temperature,
-                top_p=top_p,
-                do_sample=True,
-            ),
-            timeout=10  # 10 seconds timeout
-        )
-
-        # Extract only the AI's response
-        ai_response = response.split("AI:")[-1].strip()
-        return ai_response
-    except asyncio.TimeoutError:
-        return "I'm sorry, but I'm having trouble generating a response right now. Could you try again?"
+    response = generate_with_timeout(prompt, max_tokens, temperature, top_p)
+
+    # Extract only the AI's response
+    ai_response = response.split("AI:")[-1].strip()
+    return ai_response
 
 demo = gr.ChatInterface(
     respond,