debug

data/models/llama3-1-8b.py

@@ -24,18 +24,14 @@ pipeline = transformers.pipeline(
 @spaces.GPU(enable_queue=True)
 def chat_completion():
     data = request.json
-    
-    user_input = data.get('messages', [])
-    max_tokens = data.get('max_tokens', 2048)
-    temperature = data.get('temperature', 0.7)
-    top_p = data.get('top_p', 0.95)
 
     try:
+        print(user_input)
         outputs = pipeline(
-            user_input,
-            max_new_tokens=max_tokens, 
-            temperature=temperature,
-            top_p=top_p
+            user_input=data.get('messages', []),
+            max_new_tokens=data.get('max_tokens', 2048), 
+            temperature=data.get('temperature', 0.7),
+            top_p=data.get('top_p', 0.95)
         )
         return jsonify({"status": "success", "output": outputs[0]["generated_text"][-1]})
     except Exception as e: