add suggested_hardware

README.md

@@ -5,6 +5,7 @@ colorFrom: yellow
 colorTo: purple
 sdk: gradio
 sdk_version: 4.41.0
+suggested_hardware: a100-large
 app_port: 7860
 app_file: app.py
 models:

app.py

@@ -29,7 +29,7 @@ class StopOnTokens(StoppingCriteria):
         return False
 
 
-@spaces.GPU(duration=300)
+@spaces.GPU(duration=280)
 def predict(history, prompt, max_length, top_p, temperature):
     stop = StopOnTokens()
     messages = []
@@ -83,9 +83,9 @@ with gr.Blocks() as demo:
         </div>
         <div style="text-align: center; font-size: 15px; font-weight: bold; color: red; margin-bottom: 20px;">
             ⚠️ Due to the limitations of Huggingface ZERO GPUs, in order to output 10K characters in one go, 
-            we need to request a 5-minute quota each time. 
+            we need to request a 4-5 minute quota each time. 
             This will result in you only being able to use it once every 4 hours. 
-            If you plan to use it long-term, please consider deploying the model yourself.
+            If you plan to use it long-term, please consider deploying the model or fork this space yourself.
         </div>
         """
     )