diff --git a/vlib/toml/parser/parser.v b/vlib/toml/parser/parser.v
index e1ff7e1398..efd1ec0656 100644
--- a/vlib/toml/parser/parser.v
+++ b/vlib/toml/parser/parser.v
@@ -1127,6 +1127,13 @@ pub fn (mut p Parser) quoted() ast.Quoted {
 	mut lit := p.tok.lit[1..p.tok.lit.len - 1]
 	if is_multiline {
 		lit = p.tok.lit[3..p.tok.lit.len - 3]
+		// From https://toml.io/en/v1.0.0#string
+		// "Multi-line literal strings [...] A newline immediately following the opening
+		// delimiter will be trimmed. All other content between the delimiters
+		// is interpreted as-is without modification."
+		if lit.len > 0 && lit[0] == `\n` {
+			lit = lit[1..]
+		}
 	}
 	return ast.Quoted{
 		text: lit
diff --git a/vlib/toml/tests/burntsushi.toml-test_test.v b/vlib/toml/tests/burntsushi.toml-test_test.v
index 986d559478..8680e395e1 100644
--- a/vlib/toml/tests/burntsushi.toml-test_test.v
+++ b/vlib/toml/tests/burntsushi.toml-test_test.v
@@ -18,7 +18,8 @@ const (
 
 	valid_value_exceptions = [
 		// String
-		'string/raw-multiline.toml', // This test is not correct. Our parser *correctly* includes the newline at the start of the raw multiline.
+		'string/double-quote-escape.toml',
+		'string/unicode-escape.toml',
 		'string/escapes.toml',
 		'string/escape-tricky.toml',
 		'string/multiline.toml',
diff --git a/vlib/toml/tests/strings_test.v b/vlib/toml/tests/strings_test.v
index 8e816ad869..961c6e9cc1 100644
--- a/vlib/toml/tests/strings_test.v
+++ b/vlib/toml/tests/strings_test.v
@@ -40,9 +40,9 @@ fn test_multiline_strings() {
 	value = toml_doc.value('multi2')
 	assert value.string() == 'one\ntwo'
 	value = toml_doc.value('multi3')
-	assert value.string() == '\none\ntwo\nthree'
+	assert value.string() == 'one\ntwo\nthree'
 	value = toml_doc.value('multi4')
-	assert value.string() == '\none\ntwo\nthree\nfour\n'
+	assert value.string() == 'one\ntwo\nthree\nfour\n'
 
 	toml_doc = toml.parse(toml_multiline_text_2) or { panic(err) }
 	value = toml_doc.value('multi1')
@@ -50,9 +50,9 @@ fn test_multiline_strings() {
 	value = toml_doc.value('multi2')
 	assert value.string() == 'one\ntwo'
 	value = toml_doc.value('multi3')
-	assert value.string() == '\none\ntwo\nthree'
+	assert value.string() == 'one\ntwo\nthree'
 	value = toml_doc.value('multi4')
-	assert value.string() == '\none\ntwo\nthree\nfour\n'
+	assert value.string() == 'one\ntwo\nthree\nfour\n'
 
 	toml_file :=
 		os.real_path(os.join_path(os.dir(@FILE), 'testdata', os.file_name(@FILE).all_before_last('.'))) +
@@ -91,7 +91,7 @@ fn test_literal_strings() {
 	// See `.gitattributes` in the project root for the rule in action.
 	// These lines would look like this on Windows:
 	// assert toml_doc.value('ml_lit1').string() == '\r\n\\'
-	assert toml_doc.value('ml_lit1').string() == '\n\\'
+	assert toml_doc.value('ml_lit1').string() == '\\'
 	assert toml_doc.value('ml_lit2').string() == '\\\n\\'
 	assert toml_doc.value('ml_lit3').string() == '\\\ntricky\\\n'
 }