Update auditqa/reader.py

auditqa/reader.py

@@ -10,6 +10,7 @@ load_dotenv()
 model_config = getconfig("model_params.cfg")
 NVIDIA_SERVER = os.environ["NVIDIA_SERVERLESS"]
 HF_token = os.environ["LLAMA_3_1"]
+INF_SERVERLESS = os.environ["INF_SERVERLESS"]
 
 
 def nvidia_client():
@@ -24,15 +25,14 @@ def nvidia_client():
 
 def serverless_api():
     try:
-        HF_token = os.environ["LLAMA_3_1"]
-        if not HF_token:
+        if not INF_SERVERLESS:
             raise ValueError("LLAMA_3_1 environment variable is empty")
             
         model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
 
         client = InferenceClient(
             model=model_id,
-            api_key=HF_token,
+            api_key=INF_SERVERLESS,
         )
         
         return client