debug

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

@@ -25,10 +25,10 @@ pipeline = transformers.pipeline(
 def chat_completion():
     data = request.json
     
-    user_input = data[0].get('user_input', [])
-    max_tokens = data[0].get('max_tokens', 2048)
-    temperature = data[0].get('temperature', 0.7)
-    top_p = data[0].get('top_p', 0.95)
+    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:
         outputs = pipeline(