updating the handler

builder/requirements.txt

@@ -1,10 +1,5 @@
 # Required Python packages get listed here, one per line.
-# Reccomended to lock the version number to avoid unexpected changes.
-
-# Use the following additional index URL to get the latest PyTorch.
-# --extra-index-url https://download.pytorch.org/whl/cu118
-# vllm @ git+https://github.com/vllm-project/vllm.git@2b7d3aca2e1dd25fe26424f57c051af3b823cd71
-# runpod @ git+https://github.com/runpod/runpod-python@vllm#egg=runpod
+# Recomended to lock the version number to avoid unexpected changes.
 vllm==0.1.2
 huggingface-hub==0.16.4
 runpod @ git+https://github.com/runpod/runpod-python@main#egg=runpod

src/handler.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-''' Contains the handler function that will be called by the serverless. '''
+''' Contains the handler function that will be called by the serverless worker. '''
 
 # Start the VLLM serving layer on our RunPod worker.
 from templates import DEFAULT_TEMPLATE, LLAMA_TEMPLATE
@@ -9,8 +9,14 @@ import runpod
 import os
 
 # Prepare the model and tokenizer
-MODEL = os.environ['MODEL']
-TOKENIZER = os.environ['TOKENIZER']
+MODEL = os.environ.get('MODEL', None)
+TOKENIZER = os.environ.get('TOKENIZER', None)
+
+if not MODEL:
+    print("Error: The model has not been provided.")
+
+if not TOKENIZER:
+    print("Error: The tokenizer has not been provided.")
 
 # Prepare the engine's arguments
 engine_args = AsyncEngineArgs(