Update app.py

app.py

@@ -1,22 +1,23 @@
 import os
 from flask import Flask, request, jsonify
-from transformers import AutoTokenizer, AutoModelForCausalLM
+from transformers import pipeline
 import torch
 
-hf_token = "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"
+# Use the secret you set in Hugging Face Space settings
+hf_token = os.getenv("HF_TOKEN")
 
 MODEL_ID = "Qwen/Qwen2.5-Coder-1.0B-Instruct"
 
-print("Loading tokenizer...")
-tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, token=hf_token )
-
-print("Loading model...")
-model = AutoModelForCausalLM.from_pretrained(
-    MODEL_ID,
-    device_map="cpu",
-    token=hf_token 
+print("Loading pipeline (model + tokenizer)...")
+# The pipeline automatically downloads everything needed
+generator = pipeline(
+    "text-generation",
+    model=MODEL_ID,
+    device_map="cpu", # Force CPU for free-tier Spaces
+    torch_dtype="auto",
+    token=hf_token
 )
-print("Model loaded successfully")
+print("Pipeline loaded successfully")
 
 app = Flask(__name__)
 
@@ -25,17 +26,23 @@ def generate():
     data = request.json
     prompt = data.get("prompt", "")
     max_tokens = int(data.get("max_tokens", 256))
+
     if not prompt:
         return jsonify({"error": "Prompt required"}), 400
 
-    inputs = tokenizer(prompt, return_tensors="pt")
-    outputs = model.generate(**inputs, max_new_tokens=max_tokens)
-    text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return jsonify({"response": text})
+    # Pipeline handles tokenization, generation, and decoding automatically
+    result = generator(
+        prompt, 
+        max_new_tokens=max_tokens, 
+        truncation=True
+    )
+    
+    return jsonify({"response": result[0]['generated_text']})
 
 @app.route("/", methods=["GET"])
 def health():
     return jsonify({"status": "ok", "model": MODEL_ID})
 
 if __name__ == "__main__":
-    app.run(host="0.0.0.0", port=8000)
+    # Hugging Face Spaces require port 7860
+    app.run(host="0.0.0.0", port=7860)