dev

app.py

@@ -80,15 +80,11 @@ def generate(
     conversation.append({"role": "user", "content": message})
 
     # Use LLaMa to create chat completion
-    llm.create_chat_completion(conversation, stream=True)
+    chat_completion = llm.create_chat_completion(conversation, stream=True)
 
-    # Initialize a TextIteratorStreamer
-    streamer = TextIteratorStreamer(llm, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
-
-    outputs = []
-    for text in streamer:
-        outputs.append(text)
-        yield "".join(outputs)
+    # Yield the chat completions
+    for completion in chat_completion:
+        yield ("assistant", completion["choices"][0]["message"]["content"])
 
 chat_interface = gr.ChatInterface(
     fn=generate,