Update app.py

app.py

@@ -47,7 +47,7 @@ def stream_chat(message: str, history: list, system: str, temperature: float, ma
 
     conversation.append({"role": "user", "content": message})
 
-    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt").to(model.device)
+    input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt").to(model.device)
     streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
 
     generate_kwargs = dict(
@@ -84,7 +84,7 @@ with gr.Blocks(fill_height=True) as demo:
                 minimum=0,
                 maximum=1,
                 step=0.1,
-                value=0.95,
+                value=0.8,
                 label="Temperature",
                 render=False,
             ),