Create handler.py

handler.py

@@ -0,0 +1,23 @@
+from flask import Flask, request, jsonify
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+app = Flask(__name__)
+
+# Loading
+tokenizer = AutoTokenizer.from_pretrained("MrOvkill/Phi-3-Instruct-Bloated")
+model = AutoModelForCausalLM.from_pretrained("MrOvkill/Phi-3-Instruct-Bloated")
+
+@app.route('/predict', methods=['POST'])
+def predict():
+    data = request.json
+    prompt = data["prompt"]
+    kwargs = data.get('kwargs', {})
+    inputs = tokenizer(prompt, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(**inputs, **kwargs)
+    response = {"response": tokenizer.decode(outputs[0], skip_special_tokens=True)}
+    return jsonify(response)
+
+if __name__ == '__main__':
+    app.run()