Update main.py

main.py

@@ -4,23 +4,23 @@ import torch
 
 app = Flask(__name__)
 
-# Load model and tokenizer
-model_name = "mistralai/Mistral-7B-v0.1"
+# Use CodeLlama-7B (No authentication needed)
+model_name = "codellama/CodeLlama-7B"
 
-# Enable quantization for better performance on free-tier Spaces
+# Load model and tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
     model_name, 
-    torch_dtype=torch.float16, 
-    device_map="auto"
+    torch_dtype=torch.float16,  # Use float16 for efficiency
+    device_map="auto"           # Automatically use GPU if available
 )
 
-tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 @app.route("/")
 def home():
     return request.url
 
-@app.route("/generate")
+@app.route("/generate", methods=["POST"])
 def generate_text():
     data = request.get_json()
     prompt = data.get("prompt", "")
@@ -28,7 +28,7 @@ def generate_text():
     if not prompt:
         return jsonify({"error": "No prompt provided"}), 400
 
-    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+    inputs = tokenizer(prompt, return_tensors="pt").to("cuda" if torch.cuda.is_available() else "cpu")
     outputs = model.generate(**inputs, max_length=200)
     response_text = tokenizer.decode(outputs[0], skip_special_tokens=True)