Update chain_app.py

chain_app.py

@@ -1322,21 +1322,36 @@ async def main(message: cl.Message):
     #     ).send()
         
     elif chat_profile == 'Llama-3.1-405B':        
-        client = InferenceClient(
-            "meta-llama/Meta-Llama-3.1-405B-Instruct",
-            token=f'{hf_token_llama_3_1}',
-        )
-        
-        for message in client.chat_completion(
-            messages=[{"role": "user", "content": f'{message.content}'}],
-            max_tokens=500,
+        completion = groq_client.chat.completions.create(
+            model="llama-3.1-405b-reasoning",
+            messages=[
+                {
+                    "role": "user",
+                    "content": message.content
+                }
+            ],
+            temperature=1,
+            max_tokens=1024,
+            top_p=1,
             stream=True,
-        ):
-            complete_message += message.choiches[0].delta.content
-            await cl.Message(
-                content=complete_message,
-            ).send()
+            stop=None,
+        )
+    
+        complete_content = ""
+    
+        # Iterate over each chunk
+        for chunk in completion:
+            # Retrieve the content from the current chunk
+            content = chunk.choices[0].delta.content
+            
+            # Check if the content is not None before concatenating it
+            if content is not None:
+                complete_content += content
+    
+        # Send the concatenated content as a message
+        await cl.Message(content=complete_content).send()
 
+    
     elif chat_profile == 'Llama-3.1-70B':
         completion = groq_client.chat.completions.create(
             model="llama-3.1-70b-versatile",