Update app.py

app.py

@@ -17,7 +17,7 @@ def search(query):
     with requests.Session() as session:
         resp = session.get(
             url="https://www.google.com/search",
-            headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36"},
+            headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/111.0"},
             params={"q": term, "num": 3, "udm": 14},
             timeout=5,
             verify=None,
@@ -29,7 +29,7 @@ def search(query):
             link = result.find("a", href=True)
             link = link["href"]
             try:
-                webpage = session.get(link, headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36"}, timeout=5, verify=False)
+                webpage = session.get(link, headers={"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/111.0"}, timeout=5, verify=False)
                 webpage.raise_for_status()
                 visible_text = extract_text_from_webpage(webpage.text)
                 if len(visible_text) > max_chars_per_page:
@@ -39,7 +39,9 @@ def search(query):
                 all_results.append({"link": link, "text": None})
     return all_results
 
+# Initialize inference clients for different models
 client_gemma = InferenceClient("mistralai/Mistral-7B-Instruct-v0.3")
+client_mixtral = InferenceClient("NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO")
 client_llama = InferenceClient("meta-llama/Meta-Llama-3-8B-Instruct")
 
 func_caller = []
@@ -57,8 +59,7 @@ def respond(message, history):
         func_caller.append({"role": "user", "content": f"{str(msg[0])}"})
         func_caller.append({"role": "assistant", "content": f"{str(msg[1])}"})
 
-    message_text = message["text"]
-    func_caller.append({"role": "user", "content": f'[SYSTEM]You are a helpful assistant. You have access to the following functions: \n {str(functions_metadata)}\n\nTo use these functions respond with:\n<functioncall> {{ "name": "function_name", "arguments": {{ "arg_1": "value_1", "arg_1": "value_1", ... }} }}  </functioncall>  [USER] {message_text}'})
+    func_caller.append({"role": "user", "content": f'[SYSTEM]You are a helpful assistant. You have access to the following functions: \n {str(functions_metadata)}\n\nTo use these functions respond with:\n<functioncall> {{ "name": "function_name", "arguments": {{ "arg_1": "value_1", "arg_1": "value_1", ... }} }}  </functioncall>  [USER] {message}'})
     
     response = client_gemma.chat_completion(func_caller, max_tokens=200)
     response = str(response)
@@ -80,11 +81,11 @@ def respond(message, history):
             web_results = search(query)
             gr.Info("Extracting relevant Info")
             web2 = ' '.join([f"Link: {res['link']}\nText: {res['text']}\n\n" for res in web_results if res['text']])
-            messages = f"Web Dac uses the user agents of Mozilla, AppleWebKit, and Safari browsers for chat responses and human context mimicking."
+            messages = f"system\nYou are OpenCHAT mini a helpful assistant made by KingNish. You are provided with WEB results from which you can find informations to answer users query in Structured and More better way. You do not say Unnecesarry things Only say thing which is important and relevant. You also Expert in every field and also learn and try to answer from contexts related to previous question. Try your best to give best response possible to user. You also try to show emotions using Emojis and reply like human, use short forms, friendly tone and emotions."
             for msg in history:
                 messages += f"\nuser\n{str(msg[0])}"
                 messages += f"\nassistant\n{str(msg[1])}"
-            messages+=f"\nuser\n{message_text}\nweb_result\n{web2}\nassistant\n"
+            messages+=f"\nuser\n{message}\nweb_result\n{web2}\nassistant\n"
             stream = client_mixtral.text_generation(messages, max_new_tokens=2000, do_sample=True, stream=True, details=True, return_full_text=False)
             output = ""
             for response in stream:
@@ -92,11 +93,11 @@ def respond(message, history):
                     output += response.token.text
                     yield output
         else:
-            messages = f"Web Dac uses the user agents of Mozilla, AppleWebKit, and Safari browsers for chat responses and human context mimicking."
+            messages = f"system\nYou are OpenCHAT mini a helpful assistant made by KingNish. You answers users query like human friend. You are also Expert in every field and also learn and try to answer from contexts related to previous question. Try your best to give best response possible to user. You also try to show emotions using Emojis and reply like human, use short forms, friendly tone and emotions."
             for msg in history:
                 messages += f"\nuser\n{str(msg[0])}"
                 messages += f"\nassistant\n{str(msg[1])}"
-            messages+=f"\nuser\n{message_text}\nassistant\n"
+            messages+=f"\nuser\n{message}\nassistant\n"
             stream = client_llama.text_generation(messages, max_new_tokens=2000, do_sample=True, stream=True, details=True, return_full_text=False)
             output = ""
             for response in stream:
@@ -104,11 +105,11 @@ def respond(message, history):
                     output += response.token.text
                     yield output
     except:
-        messages = f"Web Dac uses the user agents of Mozilla, AppleWebKit, and Safari browsers for chat responses and human context mimicking."
+        messages = f"system\nYou are OpenCHAT mini a helpful assistant made by KingNish. You answers users query like human friend. You are also Expert in every field and also learn and try to answer from contexts related to previous question. Try your best to give best response possible to user. You also try to show emotions using Emojis and reply like human, use short forms, friendly tone and emotions."
         for msg in history:
             messages += f"\nuser\n{str(msg[0])}"
             messages += f"\nassistant\n{str(msg[1])}"
-        messages+=f"\nuser\n{message_text}\nassistant\n"
+        messages+=f"\nuser\n{message}\nassistant\n"
         stream = client_llama.text_generation(messages, max_new_tokens=2000, do_sample=True, stream=True, details=True, return_full_text=False)
         output = ""
         for response in stream:
@@ -120,8 +121,8 @@ demo = gr.ChatInterface(
     fn=respond,
     chatbot=gr.Chatbot(show_copy_button=True, likeable=True, layout="panel"),
     description=" ",
-    textbox=gr.Textbox(), 
-    multimodal=False, 
+    textbox=gr.Textbox(),  # Changed to Textbox
+    multimodal=False,  # Disabled multimodal
     concurrency_limit=200,
 )
-demo.launch()
+demo.launch(share=True)