diff --git a/tests/parse/test_parse_solution.py b/tests/parse/test_parse_solution.py
index 2248dafc..b959b411 100644
--- a/tests/parse/test_parse_solution.py
+++ b/tests/parse/test_parse_solution.py
@@ -24,6 +24,11 @@
             "Route #1: 1 \n Route #2: 6\n ABCDE",
             {"routes": [[1], [6]]},
         ),
+        (
+            # Reload "|" character
+            "Route #1: 1 | 2 \n Route #2: 3",
+            {"routes": [[1, "|", 2], [3]]},
+        ),
     ],
 )
 def test_parse_solution(text, data):
diff --git a/vrplib/parse/parse_solution.py b/vrplib/parse/parse_solution.py
index 3cf384f9..a21c922a 100644
--- a/vrplib/parse/parse_solution.py
+++ b/vrplib/parse/parse_solution.py
@@ -7,8 +7,11 @@
 
 def parse_solution(text: str) -> Solution:
     """
-    Parses the text of a solution file formatted in VRPLIB style. A solution
-    consists of routes, which are indexed from 1 to n, and possibly other data.
+    Parses a VRPLIB-formatted solution text into a dictionary.
+    - Routes appear as "Route #n: node1 node2 node3..." where nodes can be:
+      * Integer indices (0-indexed, unlike 1-indexed instance data)
+      * String markers (e.g., "|" for reload depots).
+    - Additional metadata as key-value pairs (e.g., "Cost: 123.45")
 
     Parameters
     ----------
@@ -18,14 +21,17 @@ def parse_solution(text: str) -> Solution:
     Returns
     -------
     dict
-        The soluion data.
+        The soluion data dictionary, containing:
+        - "routes": list of routes, each route being a list of nodes
+        - Additional metadata as key-value pairs from the solution
     """
     solution: Solution = {"routes": []}
 
     for line in text2lines(text):
         if "Route" in line:
-            route = [int(idx) for idx in line.split(":")[1].split(" ") if idx]
-            solution["routes"].append(route)  # type: ignore
+            raw_visits = line.split(":")[1].split()
+            visits = [int(val) if val.isdigit() else val for val in raw_visits]
+            solution["routes"].append(visits)  # type: ignore
         elif ":" in line or " " in line:  # Split at first colon or whitespace
             split_at = ":" if ":" in line else " "
             k, v = [word.strip() for word in line.split(split_at, 1)]