Update app.py

app.py

@@ -49,7 +49,6 @@ hf_hub_download(
 llm = None
 llm_model = None
 
-@spaces.GPU(duration=120)
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -121,7 +120,7 @@ def respond(
         outputs += output
         yield outputs
 
-description = """<p align="center">モデルとチャットテンプレートを選択してください (you can switch to other models from additional inputs)</p>
+description = """<p align="center">Defaults to Oumuamua-7b-RP (you can switch to other models from additional inputs)</p>
 <p><center>
 <a href="https://huggingface.co/Aratako/Oumuamua-7b-RP-GGUF" target="_blank">[Oumuamua-7b-RP Model]</a>
 <a href="https://huggingface.co/bartowski/Oumuamua-7b-instruct-v2-GGUF" target="_blank">[Oumuamua-7b-instruct-v2 Model]</a>
@@ -138,74 +137,40 @@ templates = [
     "B22", "LLAMA_3", "PHI_3"
 ]
 
-model_dropdown = gr.Dropdown(
-    choices=[
-        'Oumuamua-7b-RP_Q4_K_M.gguf',
-        'Oumuamua-7b-instruct-v2-Q4_K_M.gguf',
-        'umiyuki-Umievo-itr012-Gleipnir-7B-Q4_K_M.gguf',
-        'Ninja-V3-Q4_K_M.gguf',
-        'kagemusya-7b-v1Q8_0.gguf',
-        'Llama-3-ELYZA-JP-8B-q4_k_m.gguf'
-    ],
-    value="Oumuamua-7b-RP_Q4_K_M.gguf",
-    label="Model"
-)
-
-template_dropdown = gr.Dropdown(
-    choices=templates,
-    value="LLAMA_3",
-    label="Template"
-)
-
-chat_interface = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a helpful assistant.", label="System message"),
-        gr.Slider(minimum=1, maximum=4096, value=2048, step=1, label="Max tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p",
-        ),
-        gr.Slider(
-            minimum=0,
-            maximum=100,
-            value=40,
-            step=1,
-            label="Top-k",
-        ),
-        gr.Slider(
-            minimum=0.0,
-            maximum=2.0,
-            value=1.1,
-            step=0.1,
-            label="Repetition penalty",
-        ),
-    ],
-    retry_btn="Retry",
-    undo_btn="Undo",
-    clear_btn="Clear",
-    submit_btn="Send",
-    title="Chat with various models using llama.cpp", 
-    description=description,
-    chatbot=gr.Chatbot(
-        scale=1, 
-        likeable=False,
-        show_copy_button=True
-    )
-)
-
-demo = gr.Blocks()
-
-with demo:
+with gr.Blocks() as demo:
     gr.Markdown(description)
-    model_dropdown.render()
-    template_dropdown.render()
-    chat_interface.render()
+    model_dropdown = gr.Dropdown(
+        choices=[
+            'Oumuamua-7b-RP_Q4_K_M.gguf',
+            'Oumuamua-7b-instruct-v2-Q4_K_M.gguf',
+            'umiyuki-Umievo-itr012-Gleipnir-7B-Q4_K_M.gguf',
+            'Ninja-V3-Q4_K_M.gguf',
+            'kagemusya-7b-v1Q8_0.gguf',
+            'Llama-3-ELYZA-JP-8B-q4_k_m.gguf'
+        ],
+        value="Oumuamua-7b-RP_Q4_K_M.gguf",
+        label="Model"
+    )
+    template_dropdown = gr.Dropdown(
+        choices=templates,
+        value="LLAMA_3",
+        label="Template"
+    )
 
-if __name__ == "__main__":
-    demo.launch()
+    chat_interface = gr.ChatInterface(
+        fn=respond,
+        inputs=[
+            gr.Textbox(value="You are a helpful assistant.", label="System message"),
+            model_dropdown,
+            template_dropdown,
+            gr.Slider(minimum=1, maximum=4096, value=2048, step=1, label="Max tokens"),
+            gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+            gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
+            gr.Slider(minimum=0, maximum=100, value=40, step=1, label="Top-k"),
+            gr.Slider(minimum=0.0, maximum=2.0, value=1.1, step=0.1, label="Repetition penalty"),
+            gr.Chatbot(scale=1, likeable=False, show_copy_button=True)
+        ],
+        outputs="text"
+    )
 
+    chat_interface.launch()