diff --git a/.vscode/settings.json b/.vscode/settings.json
index 6d9e368f..4ef8e679 100644
--- a/.vscode/settings.json
+++ b/.vscode/settings.json
@@ -11,6 +11,7 @@
   },
   "ruby.format": "rubocop",
   "cSpell.words": [
+    "kwargs",
     "Undocumentable",
     "unshareable"
   ] // use rubocop for formatting
diff --git a/README.md b/README.md
index 44727b6d..07e11a21 100644
--- a/README.md
+++ b/README.md
@@ -346,7 +346,7 @@ The original `Crockford's base32` maps `I`, `L` to `1`, `O` to `0`.
 And accepts freestyle inserting `Hyphens (-)`.
 To consider this patterns or not is different in each implementations.
 
-Current parser/validator/matcher aims to cover `subset of Crockford's base32`.
+Current parser/validator/matcher basically aims to cover `subset of Crockford's base32`.
 I have suggested it would be clarified in [ulid/spec#57](https://github.com/ulid/spec/pull/57).
 
 >Case insensitive
@@ -356,12 +356,13 @@ But it is a controversial point, discussing in [ulid/spec#3](https://github.com/
 
 Be that as it may, this gem provides API for handling the nasty possibilities.
 
-`ULID.normalize` and `ULID.normalized?`
+`ULID.normalize`, `ULID.normalized?`, `ULID.valid_as_variants?`
 
 ```ruby
-ULID.normalize('-olarz3-noekisv4rrff-q6ig5fav--') #=> "01ARZ3N0EK1SV4RRFFQ61G5FAV"
-ULID.normalized?('-olarz3-noekisv4rrff-q6ig5fav--') #=> false
-ULID.normalized?('01ARZ3N0EK1SV4RRFFQ61G5FAV') #=> true
+ULID.normalize('01g70y0y7g-z1xwdarexergsddd') #=> "01G70Y0Y7GZ1XWDAREXERGSDDD"
+ULID.normalized?('01g70y0y7g-z1xwdarexergsddd') #=> false
+ULID.normalized?('01G70Y0Y7GZ1XWDAREXERGSDDD') #=> true
+ULID.valid_as_variants?('01g70y0y7g-z1xwdarexergsddd') #=> true
 ```
 
 #### UUIDv4 converter (experimental)
diff --git a/lib/ulid.rb b/lib/ulid.rb
index 592a3198..195dea5a 100644
--- a/lib/ulid.rb
+++ b/lib/ulid.rb
@@ -280,18 +280,36 @@ def self.normalize(string)
     parse(normalized_in_crockford).to_s
   end
 
+  # @param [String, #to_str] string
   # @return [Boolean]
-  def self.normalized?(object)
-    normalized = normalize(object)
+  def self.normalized?(string)
+    normalized = normalize(string)
   rescue Exception
     false
   else
-    normalized == object
+    normalized == string
   end
 
+  # @param [String, #to_str] string
   # @return [Boolean]
-  def self.valid?(object)
-    string = String.try_convert(object)
+  def self.valid_as_variants?(string)
+    normalize(string)
+  rescue Exception
+    false
+  else
+    true
+  end
+
+  # @deprecated Use [.valid_as_variants?] or [.normalized?] instead
+  #
+  # Returns `true` if it is normalized string.
+  # Basically the difference of normalized? is to accept downcase or not. This returns true for downcased ULIDs.
+  #
+  # @return [Boolean]
+  def self.valid?(string)
+    warn_kwargs = (RUBY_VERSION >= '3.0') ? { category: :deprecated } : {}
+    Warning.warn('ULID.valid? is deprecated. Use ULID.valid_as_variants? or ULID.normalized? instead.', **warn_kwargs)
+    string = String.try_convert(string)
     string ? STRICT_PATTERN_WITH_CROCKFORD_BASE32_SUBSET.match?(string) : false
   end
 
diff --git a/sig/ulid.rbs b/sig/ulid.rbs
index b0656a4d..72e0257d 100644
--- a/sig/ulid.rbs
+++ b/sig/ulid.rbs
@@ -309,14 +309,13 @@ class ULID < Object
   # ```
   def self.sample: (?period: period) -> ULID
                  | (Integer number, ?period: period?) -> Array[ULID]
-  def self.valid?: (untyped) -> bool
 
   # Returns normalized string
   #
   # ```ruby
+  # ULID.normalize('01G70Y0Y7G-Z1XWDAREXERGSDDD') #=> "01G70Y0Y7GZ1XWDAREXERGSDDD"
   # ULID.normalize('-olarz3-noekisv4rrff-q6ig5fav--') #=> "01ARZ3N0EK1SV4RRFFQ61G5FAV"
-  # ULID.normalized?('-olarz3-noekisv4rrff-q6ig5fav--') #=> false
-  # ULID.normalized?('01ARZ3N0EK1SV4RRFFQ61G5FAV') #=> true
+  # ULID.normalize('01G70Y0Y7G_Z1XWDAREXERGSDDD') #=> ULID::ParserError
   # ```
   #
   # See also [ulid/spec#57](https://github.com/ulid/spec/pull/57) and [ulid/spec#3](https://github.com/ulid/spec/issues/3)
@@ -325,13 +324,40 @@ class ULID < Object
   # Returns `true` if it is normalized string
   #
   # ```ruby
-  # ULID.normalize('-olarz3-noekisv4rrff-q6ig5fav--') #=> "01ARZ3N0EK1SV4RRFFQ61G5FAV"
-  # ULID.normalized?('-olarz3-noekisv4rrff-q6ig5fav--') #=> false
-  # ULID.normalized?('01ARZ3N0EK1SV4RRFFQ61G5FAV') #=> true
+  # ULID.normalized?('01G70Y0Y7GZ1XWDAREXERGSDDD') #=> true
+  # ULID.normalized?('01G70Y0Y7G-Z1XWDAREXERGSDDD') #=> false
+  # ULID.normalized?(ULID.generate.to_s.downcase) #=> false
+  # ULID.normalized?('01G70Y0Y7G_Z1XWDAREXERGSDDD') #=> false (Not raising ULID::ParserError)
+  # ```
+  #
+  # See also [ulid/spec#57](https://github.com/ulid/spec/pull/57) and [ulid/spec#3](https://github.com/ulid/spec/issues/3)
+  def self.normalized?: (_ToStr string) -> bool
+                      | (untyped) -> false
+
+  # Returns `true` if it is valid in ULID format variants
+  #
+  # ```ruby
+  # ULID.valid_as_variants?(ULID.generate.to_s.downcase) #=> true
+  # ULID.valid_as_variants?('01G70Y0Y7G-Z1XWDAREXERGSDDD') #=> true
+  # ULID.valid_as_variants?('01G70Y0Y7G_Z1XWDAREXERGSDDD') #=> false
+  # ```
+  #
+  # See also [ulid/spec#57](https://github.com/ulid/spec/pull/57) and [ulid/spec#3](https://github.com/ulid/spec/issues/3)
+  def self.valid_as_variants?: (_ToStr string) -> bool
+                             | (untyped) -> false
+
+  # DEPRECATED Use valid_as_variants? instead
+  #
+  # Returns `true` if it is normalized string.
+  # Basically the difference of normalized? is to accept downcase or not. This returns true for downcased ULIDs.
+  #
+  # ```ruby
+  # ULID.valid?(ULID.generate.to_s.downcase) #=> true
   # ```
   #
   # See also [ulid/spec#57](https://github.com/ulid/spec/pull/57) and [ulid/spec#3](https://github.com/ulid/spec/issues/3)
-  def self.normalized?: (untyped) -> bool
+  def self.valid?: (_ToStr string) -> bool
+                 | (untyped) -> false
 
   # Returns parsed ULIDs from given String for rough operations.
   #
diff --git a/steep_expectations.yml b/steep_expectations.yml
index dcd024e9..6b946e6f 100644
--- a/steep_expectations.yml
+++ b/steep_expectations.yml
@@ -1 +1,13 @@
---- []
+---
+- file: lib/ulid.rb
+  diagnostics:
+  - range:
+      start:
+        line: 311
+        character: 12
+      end:
+        line: 311
+        character: 16
+    severity: ERROR
+    message: Type `singleton(::Warning)` does not have method `warn`
+    code: Ruby::NoMethod
diff --git a/test/core/test_ulid_class.rb b/test/core/test_ulid_class.rb
index dc7876c3..b7508d88 100644
--- a/test/core/test_ulid_class.rb
+++ b/test/core/test_ulid_class.rb
@@ -36,7 +36,8 @@ def test_exposed_methods
         :range,
         :at,
         :normalized?,
-        :parse
+        :parse,
+        :valid_as_variants?
       ].sort,
       exposed_methods.sort
     )
@@ -136,19 +137,23 @@ def test_from_milliseconds_and_entropy
   end
 
   def test_valid?
-    assert_equal(false, ULID.valid?(nil))
-    assert_equal(false, ULID.valid?(''))
-    assert_equal(false, ULID.valid?(BasicObject.new))
-    assert_equal(false, ULID.valid?(Object.new))
-    assert_equal(false, ULID.valid?(42))
-    assert_equal(false, ULID.valid?(:'01ARZ3NDEKTSV4RRFFQ69G5FAV'))
-    assert_equal(false, ULID.valid?(ULID.sample))
-    assert_equal(false, ULID.valid?("01ARZ3NDEKTSV4RRFFQ69G5FAV\n"))
-    assert_equal(false, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAU'))
-    assert_equal(true, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAV'))
-    assert_equal(true, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAV'.downcase))
-    assert_equal(true, ULID.valid?('7ZZZZZZZZZZZZZZZZZZZZZZZZZ'))
-    assert_equal(false, ULID.valid?('80000000000000000000000000'))
+    assert_warning('ULID.valid? is deprecated. Use ULID.valid_as_variants? or ULID.normalized? instead.') do
+      assert_equal(false, ULID.valid?(nil))
+      assert_equal(false, ULID.valid?(''))
+      assert_equal(false, ULID.valid?(BasicObject.new))
+      assert_equal(false, ULID.valid?(Object.new))
+      assert_equal(false, ULID.valid?(42))
+      assert_equal(false, ULID.valid?(:'01ARZ3NDEKTSV4RRFFQ69G5FAV'))
+      assert_equal(false, ULID.valid?(ULID.sample))
+      assert_equal(false, ULID.valid?("01ARZ3NDEKTSV4RRFFQ69G5FAV\n"))
+      assert_equal(false, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAU'))
+      assert_equal(true, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAV'))
+      assert_equal(true, ULID.valid?('01ARZ3NDEKTSV4RRFFQ69G5FAV'.downcase))
+      assert_equal(true, ULID.valid?('7ZZZZZZZZZZZZZZZZZZZZZZZZZ'))
+      assert_equal(false, ULID.valid?('80000000000000000000000000'))
+
+      assert_false(ULID.valid?('01G70Y0Y7G-Z1XWDAREXERGSDDD'))
+    end
 
     assert_raises(ArgumentError) do
       ULID.valid?
@@ -215,6 +220,8 @@ def test_normalize
   end
 
   def test_normalized?
+    assert_false(ULID.normalized?('01G70Y0Y7G-Z1XWDAREXERGSDDD'))
+
     nasty = '-olarz3-noekisv4rrff-q6ig5fav--'
     assert_equal(false, ULID.normalized?(nasty))
     assert_equal(true, ULID.normalized?(ULID.normalize(nasty)))
@@ -247,6 +254,41 @@ def test_normalized?
     end
   end
 
+  def test_valid_as_variants?
+    assert_true(ULID.valid_as_variants?('01G70Y0Y7G-Z1XWDAREXERGSDDD'))
+
+    nasty = '-olarz3-noekisv4rrff-q6ig5fav--'
+    assert_true(ULID.valid_as_variants?(nasty))
+    assert_true(ULID.valid_as_variants?(ULID.normalize(nasty)))
+
+    normalized = '01ARZ3NDEKTSV4RRFFQ69G5FAV'
+    assert_true(ULID.valid_as_variants?(normalized))
+    assert_true(ULID.valid_as_variants?(normalized.downcase))
+
+    [
+      '',
+      "01ARZ3NDEKTSV4RRFFQ69G5FAV\n",
+      '01ARZ3NDEKTSV4RRFFQ69G5FAU',
+      '01ARZ3NDEKTSV4RRFFQ69G5FA',
+      '80000000000000000000000000'
+    ].each do |invalid|
+      assert_false(ULID.valid_as_variants?(invalid))
+    end
+
+    ULID.sample(1000).each do |sample|
+      assert_true(ULID.valid_as_variants?(sample.to_s))
+      assert_true(ULID.valid_as_variants?(sample.to_s.downcase))
+    end
+
+    assert_raises(ArgumentError) do
+      ULID.valid_as_variants?
+    end
+
+    [nil, 42, normalized.to_sym, BasicObject.new, Object.new, ULID.parse(normalized)].each do |evil|
+      assert_false(ULID.valid_as_variants?(evil))
+    end
+  end
+
   def test_range
     time_has_more_value_than_milliseconds1 = Time.at(946684800, Rational('123456.789')) # 2000-01-01 00:00:00.123456789 UTC
     time_has_more_value_than_milliseconds2 = Time.at(1620045632, Rational('123456.789')) # 2021-05-03 12:40:32.123456789 UTC
diff --git a/test/helper.rb b/test/helper.rb
index 385cc21c..37927791 100644
--- a/test/helper.rb
+++ b/test/helper.rb
@@ -1,20 +1,27 @@
 # coding: us-ascii
 # frozen_string_literal: true
 
+# How to use: https://github.com/jeremyevans/ruby-warning
 require('warning')
 
-# How to use => https://test-unit.github.io/test-unit/en/
+# How to use: https://test-unit.github.io/test-unit/en/
 require('test/unit')
 
 require('irb')
 require('power_assert/colorize')
 require('irb/power_assert')
 
+require 'stringio'
+
 Warning[:deprecated] = true
 Warning[:experimental] = true
 
-Warning.process do |_warning|
-  :raise
+Warning.process do |warning_message|
+  if /ULID.valid\? is deprecated/.match?(warning_message)
+    :default
+  else
+    :raise
+  end
 end
 
 require_relative('../lib/ulid')
@@ -37,4 +44,17 @@ def assert_acceptable_randomized_string(ulids)
       assert_in_epsilon(awesome_randomized_ulids.size, ulids.size, (5/100r).to_f)
     end
   end
+
+  def assert_warning(pattern, &block)
+    org_stderr = $stderr
+    $stderr = fake_io = StringIO.new(+'', 'r+')
+
+    begin
+      block.call
+      fake_io.rewind
+      assert_match(pattern, fake_io.read)
+    ensure
+      $stderr = org_stderr
+    end
+  end
 end
diff --git a/test/many_data/test_fixed_many_data.rb b/test/many_data/test_fixed_many_data.rb
index 15bb4d4d..5b977bb4 100644
--- a/test/many_data/test_fixed_many_data.rb
+++ b/test/many_data/test_fixed_many_data.rb
@@ -25,7 +25,7 @@ def assert_example(ulid, example)
     assert_equal(example.octets, ulid.octets)
 
     assert do
-      ULID.valid?(example.string)
+      ULID.normalized?(example.string)
     end
   end