diff --git a/multibase/converters.py b/multibase/converters.py
index 4202658..ca30c4b 100644
--- a/multibase/converters.py
+++ b/multibase/converters.py
@@ -155,7 +155,12 @@ def encode(self, bytes):
         return self._encode_bytes(ensure_bytes(bytes), 5, 8, 5, 8)
 
     def decode(self, bytes):
-        return self._decode_bytes(ensure_bytes(bytes), 8, 5, 8)
+        data = ensure_bytes(bytes)
+        if self.digits.islower():
+            data = data.lower()
+        elif self.digits.isupper():
+            data = data.upper()
+        return self._decode_bytes(data, 8, 5, 8)
 
 
 class Base256EmojiConverter:
diff --git a/tests/test_multibase.py b/tests/test_multibase.py
index 77bad99..9ea9022 100644
--- a/tests/test_multibase.py
+++ b/tests/test_multibase.py
@@ -128,6 +128,20 @@ def test_decode(_, data, encoded_data):
     assert decode(encoded_data) == ensure_bytes(data)
 
 
+def test_decode_base32_case_insensitive():
+    # base32 (prefix 'b', lowercase alphabet), decoding uppercase payload
+    assert decode("bMZXW6") == b"foo"
+
+    # base32upper (prefix 'B', uppercase alphabet), decoding lowercase payload
+    assert decode("Bmzxw6") == b"foo"
+
+    # base32pad (prefix 'c', lowercase alphabet), decoding uppercase payload
+    assert decode("cMZXW6====") == b"foo"
+
+    # base32hex (prefix 'v', lowercase alphabet), decoding uppercase payload
+    assert decode("vCPNMU") == b"foo"
+
+
 @pytest.mark.parametrize("encoded_data", INCORRECT_ENCODED_DATA)
 def test_decode_incorrect_encoding(encoded_data):
     with pytest.raises(InvalidMultibaseStringError) as excinfo: