updating respond function

app.py

@@ -57,15 +57,21 @@ def respond(message, history, system_message, max_tokens, temperature, top_p):
         prompt = f"{system_message}\nUser: {message}\nAnswer:"
         response = ""
 
-        for chunk in client.text2text_generation(
-            prompt,
-            max_new_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-            stream=True,
+        for chunk in client.post(
+            json={
+                "inputs": prompt,
+                "parameters": {
+                    "max_new_tokens": max_tokens,
+                    "temperature": temperature,
+                    "top_p": top_p,
+                },
+                "stream": True,
+            }
         ):
-            response += chunk
-            yield response.strip()
+            if "token" in chunk:
+                token = chunk["token"]["text"]
+                response += token
+                yield response.strip()
 
         if not response.strip():
             yield "⚠️ No response from model (check model ID or token)."
@@ -75,6 +81,7 @@ def respond(message, history, system_message, max_tokens, temperature, top_p):
         yield f"⚠️ Error: {str(e)}\n\nTraceback:\n{tb}"
 
 
+
 chatbot = gr.ChatInterface(
     fn=respond,
     additional_inputs=[