Update app.py

app.py

@@ -1,3 +1,17 @@
+from transformers import pipeline,LlamaForCausalLM,AutoTokenizer
 import gradio as gr
+import spaces
+# ignore_mismatched_sizes=True
+tokenizer = AutoTokenizer.from_pretrained('models/google/gemma-2-9b')
+model = LlamaForCausalLM.from_pretrained('models/google/gemma-2-9b',ignore_mismatched_sizes=True)
+pipe = pipeline('text-generation', model=model,tokenizer = tokenizer)
 
-gr.load("models/google/gemma-2-9b").launch()
+@spaces.GPU(duration=120)
+def generate(prompt):
+    return pipe(prompt)
+
+gr.Interface(
+    fn=generate,
+    inputs=gr.Text(),
+    outputs="text",
+).launch()