fixed chatbot

app.py

@@ -57,8 +57,8 @@ with demo:
             This function simulates streaming by yielding characters one by one.
             """
             url = inference_endpoint_url
-            params = {"query": query,"selected_model":chat_model}
-            with requests.get(url, json=params, stream=True) as r:
+            params = {"query": query, "selected_model": chat_model}
+            with requests.get(url, json=params, stream=True) as r:  # Use params for query parameters
                 for chunk in r.iter_content(chunk_size=1):
                     if chunk:
                         yield chunk.decode()
@@ -68,16 +68,38 @@ with demo:
             Wrapper function to call the streaming API and compile the response.
             """
             response = ''
-            
-            global chat_model_selection
-            
             for char in call_api_and_stream_response(query, chat_model=chat_model_selection):
-                if char == '<':
+                if char == '<':  # This seems to be your stopping condition; adjust as needed.
                     break
                 response += char
-                yield response
-        
-        gr.ChatInterface(get_response, retry_btn = None, undo_btn=None, concurrency_limit=inference_concurrency_limit).launch()   
+                yield [(f"🤖 Response from LLM: {chat_model_selection}", response)]  # Correct format for Gradio Chatbot
+
+        with gr.Blocks() as chat_interface:
+            chatbot = gr.Chatbot()
+            msg = gr.Textbox()
+            submit = gr.Button("Submit")
+            clear = gr.Button("Clear")
+
+            def user(user_message, history):
+                return "", history + [[user_message, None]]
+
+            def clear_chat(*args):
+                return []  # Returning an empty list to signify clearing the chat, adjust as per Gradio's capabilities
+
+            submit.click(
+                fn=get_response,
+                inputs=[msg, chatbot],
+                outputs=chatbot
+            )
+
+            clear.click(
+                fn=clear_chat,
+                inputs=None,
+                outputs=chatbot
+            )
+            
+        chat_interface.queue()
+        chat_interface.launch()
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
         with gr.TabItem("🏆 LLM Leadeboard", elem_id="llm-benchmark-table", id=0):