Update app.py

app.py

@@ -102,9 +102,14 @@ def respond(
     stream = llm.create_chat_completion(messages=messages,temperature = temperature,top_k = top_k,top_p = top_p,min_p = min_p,max_tokens = max_tokens,repeat_penalty = repeat_penalty,stream = True)
     
     outputs = ""
-    for output in stream:
-        outputs += output
-        yield outputs.replace("<|assistant|>","").replace("<|user|>","")
+    for chunk in stream:
+        delta = chunk['choices'][0]['delta']
+        if 'content' in delta:
+            tokens = delta['content'].split()
+            for token in tokens:
+                yield token
+            
+        #yield outputs.replace("<|assistant|>","").replace("<|user|>","")
 
 description = """
 """