Update app.py

app.py

@@ -11,6 +11,7 @@ def load_model(model_name):
 @spaces.GPU()
 def generate(
     model_name,
+    system,
     user_input,
     temperature=0.4,
     top_p=0.95,
@@ -27,7 +28,7 @@ def generate(
     if model_name == "M4-ai/tau-1.8B":
         prompt = user_input
     else:
-        prompt = f"<|im_start|>user\n{user_input}<|im_end|>\n<|im_start|>assistant\n"
+        prompt = f"<|im_start|>system\n{system}<|im_end|>\n<|im_start|>user\n{user_input}<|im_end|>\n<|im_start|>assistant\n"
     streamer = TextIteratorStreamer(pipe.tokenizer, timeout=240.0, skip_prompt=True, skip_special_tokens=True)
     generation_kwargs = dict(text_inputs=prompt, streamer=streamer, max_new_tokens=max_new_tokens, do_sample=True, top_p=top_p, min_p=min_p, top_k=top_k, 
                               temperature=temperature, num_beams=1, repetition_penalty=1.1)
@@ -44,6 +45,7 @@ g = gr.Interface(
     fn=generate,
     inputs=[
         gr.components.Dropdown(choices=model_choices, label="Model", value=model_choices[0], interactive=True),
+        gr.components.Textbox(lines=2, label="System Prompt", value="You are an AI."),
         gr.components.Textbox(lines=2, label="Prompt", value="Write me a Python program that calculates the factorial of a given number."),
         gr.components.Slider(minimum=0, maximum=1, value=0.8, label="Temperature"),
         gr.components.Slider(minimum=0, maximum=1, value=0.95, label="Top p"),