Putting the pipe outside - Trial

app.py

@@ -7,6 +7,9 @@ import torch
 # Use a pipeline as a high-level helper
 from transformers import pipeline
 
+print("RUNNING PIPE")
+pipe = pipeline("text-generation", model="NousResearch/Hermes-3-Llama-3.1-8B", max_new_tokens=200, device=0)
+
 
 @spaces.GPU(duration=120)
 def llama3_1_8B(question):
@@ -22,9 +25,7 @@ def llama3_1_8B(question):
                     
     else:
         print("CUDA is not available.")
-    print("RUNNING PIPE")
-    pipe = pipeline("text-generation", model="NousResearch/Hermes-3-Llama-3.1-8B", max_new_tokens=200, device=0)
-    
+
     print("GATHERING RESPONSES")
     responses = pipe(messages)