first

app.py

@@ -2,13 +2,17 @@ import gradio as gr
 import spaces
 import torch
 
+import transformers
+import torch
+
+model_id = "meta-llama/Meta-Llama-3-8B"
 
+pipeline = transformers.pipeline(
+    "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto"
+)
 @spaces.GPU
 def yes_man(message, history):
-    if message.endswith("?"):
-        return "Yes"
-    else:
-        return "Ask me anything!"
+    return pipeline(message)
 
 gr.ChatInterface(
     yes_man,