Update app.py

app.py

@@ -141,7 +141,7 @@ def initialize_model():
     model = AutoModelForCausalLM.from_pretrained(
         MODEL_ID,
         device_map="cuda",
-        quantization_config=quantization_config,
+        #quantization_config=quantization_config,
         torch_dtype=torch.bfloat16,
         trust_remote_code=True
     ).to("cuda")
@@ -213,7 +213,7 @@ with gr.Blocks(css=CSS, theme=gr.themes.Soft()) as demo:
     
     with gr.Accordion("⚙️ Settings", open=False):
         system_prompt = gr.TextArea(value=DEFAULT_SYSTEM_PROMPT, label="System Instructions")
-        temperature = gr.Slider(0, 1, value=0.6, label="Creativity")
+        temperature = gr.Slider(0, 1, value=0.3, label="Creativity")
         max_tokens = gr.Slider(128, 8192, value=2048, label="Max Response Length")
 
     clear = gr.Button("Clear History")