Update app.py

app.py

@@ -80,23 +80,15 @@ def get_top_chunks(query, chunk_embeddings, text_chunks):
   # Return the list of most relevant chunks
   return top_chunks
 
-client = InferenceClient("microsoft/phi-4")
-
-def respond (message, history):
-    information = get_top_chunks(message,chunk_embeddings,cleaned_chunks)
-    
-    messages = [{"role": "system", "content":"You are a friendly chatbot."}]
+client = InferenceClient("Qwen/Qwen2.5-7B-Instruct-1M")
 
+def respond(message, history):
+    messages = [{"role":"system", "content": "You are a friendly chatbot."}]
     if history:
         messages.extend(history)
-
     messages.append({"role": "user", "content": message})
-
-    response = client.chat_completion(
-        messages, max_tokens = 150, temperature = 0.2
-    )
-    
-    return response
+    response = client.chat_completion(messages,max_tokens=100)
+    return response["choices"][0]["message"]["content"].strip()
 
 chatBot = gr.ChatInterface(respond, type = "messages")
 chatBot.launch()