Update app.py

app.py

@@ -30,19 +30,14 @@ def load_model(model_id):
 if not torch.cuda.is_available():
     DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
 
-if torch.cuda.is_available():
-    model_id = "Nekochu/Luminia-13B-v3"
-    model, tokenizer = load_model(model_id)
-
 MODELS = [
-    {"name": "Nekochu/Luminia-13B-v3", "id": "Nekochu/Luminia-13B-v3"},
-    {"name": "Nekochu/Llama-2-13B-German-ORPO", "id": "Nekochu/Llama-2-13B-German-ORPO"},
+    "Nekochu/Luminia-13B-v3",
+    "Nekochu/Llama-2-13B-German-ORPO",
 ]
 
 @spaces.GPU(duration=120)
 def generate(
-    model_dropdown: str,
-    custom_model_id: str,
+    model_id: str,
     message: str,
     chat_history: list[tuple[str, str]],
     system_prompt: str,
@@ -52,8 +47,7 @@ def generate(
     top_k: int = 50,
     repetition_penalty: float = 1.2,
 ) -> Iterator[str]:
-    selected_model_id = custom_model_id if custom_model_id else model_dropdown
-    model, tokenizer = load_model(selected_model_id)
+    model, tokenizer = load_model(model_id)
 
     conversation = []
     if system_prompt:
@@ -88,18 +82,10 @@ def generate(
         outputs.append(text)
         yield "".join(outputs)
 
-model_dropdown = gr.Dropdown(
-    label="Select Predefined Model",
-    choices=[model["name"] for model in MODELS],
-    value=MODELS[0]["name"], # Default to the first model
-)
-custom_model_id_input = gr.Textbox(label="Or Enter Custom Model ID", placeholder="Enter model ID here")
-
 chat_interface = gr.ChatInterface(
     fn=generate,
     additional_inputs=[
-        model_dropdown,
-        custom_model_id_input,
+        gr.Dropdown(label="Select Model", choices=MODELS),
         gr.Textbox(label="System prompt", lines=6),
         gr.Slider(
             label="Max new tokens",
@@ -151,4 +137,4 @@ with gr.Blocks(css="style.css") as demo:
     gr.Markdown(LICENSE)
 
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
+    demo.queue(max_size=20).launch()