Update auditqa/reader.py

auditqa/reader.py

@@ -21,6 +21,26 @@ def nvidia_client():
 
     return client
 
+
+def serverless_api():
+    try:
+        HF_token = os.environ["LLAMA_3_1"]
+        if not HF_token:
+            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,
+        )
+        
+        return client
+
+    except Exception as e:
+        raise
+
+
 def dedicated_endpoint():
     """ returns the dedicated server endpoint"""