diff --git a/.gitignore b/.gitignore
index e18b131..6c16453 100644
--- a/.gitignore
+++ b/.gitignore
@@ -155,6 +155,7 @@ cython_debug/
 
 #ML
 VeraMind-Mini/
+Token.txt
 
 # OS generated files #
 ######################
diff --git a/src/Ai/llm.py b/src/Ai/llm.py
new file mode 100644
index 0000000..8343f19
--- /dev/null
+++ b/src/Ai/llm.py
@@ -0,0 +1,40 @@
+from langchain_community.llms import Ollama
+
+class ArticleRater:
+    def __init__(self):
+        self.client = "https://ai.fabelous.app/v1/ollama/generic"
+        self.headers = self._load_token("Token.txt")
+
+    def _load_token(self, token_path):
+        with open(token_path, "r") as f:
+            token = f.read().strip()
+        return {"Authorization": f"Token {token}"}
+
+    def get_response(self, article, result):
+        ollama_params = {
+            "base_url": self.client,
+            "model": "mistral-nemo:12b-instruct-2407-q8_0",
+            "headers": self.headers,
+            "system": "Give a short explanation max 1-3 sentence why this article is rated like that"
+        }
+
+        message = (f"A Machine Learning Model labeled an article the following: "
+                   f"Result: {result['result']}, Confidence: {result['confidence']}, The Article: {article}")
+
+        # Initialize the Ollama object with the prepared parameters
+        llm = Ollama(**ollama_params)
+
+        # Return the response stream
+        return llm.stream(message)
+
+# Usage
+if __name__ == "main":
+    article_rater = ArticleRater()
+
+    article = "Example article content."
+    result = {"result": "REAL", "confidence": 0.754}
+
+    # Capture the stream response
+    response_stream = article_rater.get_response(article, result)
+    for chunk in response_stream:
+        print(chunk, end='', flush=True)