feat: add

app.py

@@ -7,14 +7,18 @@ def greet(name):
 
 
 @spaces.GPU
-def infer(name: str = "Who are you?"):
-    from transformers import pipeline
+def infer(input_text: str = "Who are you?"):
+
+    # messages = [
+    #     {"role": "user", "content": name},
+    # ]
+    model = transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
+    token = transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
+    token.encode(input_text, return_tensors="pt" )
+    output = model(input_ids)
+    print(output)
+    return output
 
-    messages = [
-        {"role": "user", "content": name},
-    ]
-    pipe = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
-    return pipe(messages)
 
 text_input = gr.Textbox(label="Input Text", placeholder="test")