diff --git a/llm_server/llm/info.py b/llm_server/llm/info.py
index 907e10e..f9a5369 100644
--- a/llm_server/llm/info.py
+++ b/llm_server/llm/info.py
@@ -8,21 +8,21 @@ def get_running_model():
         try:
             backend_response = requests.get(f'{opts.backend_url}/api/v1/model')
         except Exception as e:
-            return False
+            return False, e
         try:
             r_json = backend_response.json()
             return r_json['result']
         except Exception as e:
-            return False
+            return False, e
     elif opts.mode == 'hf-textgen':
         try:
             backend_response = requests.get(f'{opts.backend_url}/info')
         except Exception as e:
-            return False
+            return False, e
         try:
             r_json = backend_response.json()
             return r_json['model_id'].replace('/', '_')
         except Exception as e:
-            return False
+            return False, e
     else:
         raise Exception
diff --git a/llm_server/routes/v1/info.py b/llm_server/routes/v1/info.py
index 3ffea16..ce15154 100644
--- a/llm_server/routes/v1/info.py
+++ b/llm_server/routes/v1/info.py
@@ -22,11 +22,12 @@ from ..cache import cache
 @bp.route('/model', methods=['GET'])
 @cache.cached(timeout=60, query_string=True)
 def get_model():
-    model = get_running_model()
+    model, error = get_running_model()
     if not model:
         return jsonify({
             'code': 500,
-            'error': 'failed to reach backend'
+            'error': 'failed to reach backend',
+            'msg': error
         }), 500
     else:
         return jsonify({