diff --git a/lib/parse.js b/lib/parse.js
index 7d343ca..480c2ad 100644
--- a/lib/parse.js
+++ b/lib/parse.js
@@ -12,6 +12,14 @@ var uUpper = "U".charCodeAt(0);
 var plus = "+".charCodeAt(0);
 var isUnicodeRange = /^[a-f0-9?-]+$/i;
 
+function isHexDigit(code) {
+  return (
+    (code >= 48 && code <= 57) ||
+    (code >= 65 && code <= 70) ||
+    (code >= 97 && code <= 102)
+  );
+}
+
 module.exports = function (input) {
   var tokens = [];
   var value = input;
@@ -261,6 +269,29 @@ module.exports = function (input) {
       do {
         if (code === backslash) {
           next += 1;
+          code = value.charCodeAt(next);
+          if (isHexDigit(code)) {
+            escapePos = next;
+            do {
+              next += 1;
+              code = value.charCodeAt(next);
+            } while (next - escapePos < 6 && isHexDigit(code));
+            // A hex escape can end in one CSS whitespace character.
+            if (
+              code === 32 ||
+              code === 9 ||
+              code === 10 ||
+              code === 12 ||
+              code === 13
+            ) {
+              next += 1;
+              if (code === 13 && value.charCodeAt(next) === 10) {
+                next += 1;
+              }
+            }
+            code = value.charCodeAt(next);
+            continue;
+          }
         }
         next += 1;
         code = value.charCodeAt(next);
diff --git a/test/hex-escapes.js b/test/hex-escapes.js
new file mode 100644
index 0000000..dec0ce4
--- /dev/null
+++ b/test/hex-escapes.js
@@ -0,0 +1,46 @@
+const { test } = require("node:test");
+const { deepEqual, strictEqual } = require("node:assert");
+const valueParser = require("../lib");
+
+for (const whitespace of [" ", "\t", "\n", "\r", "\f", "\r\n"]) {
+  test(`hex escape consumes its whitespace terminator ${JSON.stringify(whitespace)}`, () => {
+    const value = "fo\\a" + whitespace + "o";
+    const parsed = valueParser(value);
+    deepEqual(parsed.nodes, [
+      { type: "word", sourceIndex: 0, sourceEndIndex: value.length, value },
+    ]);
+    strictEqual(parsed.toString(), value);
+  });
+}
+
+test("hex escape consumes only one whitespace terminator", () => {
+  deepEqual(valueParser("fo\\a  o").nodes, [
+    { type: "word", sourceIndex: 0, sourceEndIndex: 5, value: "fo\\a " },
+    { type: "space", sourceIndex: 5, sourceEndIndex: 6, value: " " },
+    { type: "word", sourceIndex: 6, sourceEndIndex: 7, value: "o" },
+  ]);
+});
+
+test("hex escape stops after six digits", () => {
+  const value = "\\123456a o";
+  deepEqual(valueParser(value).nodes, [
+    { type: "word", sourceIndex: 0, sourceEndIndex: 8, value: "\\123456a" },
+    { type: "space", sourceIndex: 8, sourceEndIndex: 9, value: " " },
+    { type: "word", sourceIndex: 9, sourceEndIndex: 10, value: "o" },
+  ]);
+});
+
+test("hex escape preserves a following divider", () => {
+  deepEqual(valueParser("\\41 ,x").nodes, [
+    { type: "word", sourceIndex: 0, sourceEndIndex: 4, value: "\\41 " },
+    {
+      type: "div",
+      sourceIndex: 4,
+      sourceEndIndex: 5,
+      value: ",",
+      before: "",
+      after: "",
+    },
+    { type: "word", sourceIndex: 5, sourceEndIndex: 6, value: "x" },
+  ]);
+});
diff --git a/test/parse.js b/test/parse.js
index 22aa393..a6950ca 100644
--- a/test/parse.js
+++ b/test/parse.js
@@ -1559,9 +1559,7 @@ const tests = [
     expected: [
       { type: "word", sourceIndex: 0, sourceEndIndex: 6, value: "Hawaii" },
       { type: "space", sourceIndex: 6, sourceEndIndex: 7, value: " " },
-      { type: "word", sourceIndex: 7, sourceEndIndex: 10, value: "\\35" },
-      { type: "space", sourceIndex: 10, sourceEndIndex: 11, value: " " },
-      { type: "word", sourceIndex: 11, sourceEndIndex: 13, value: "-0" },
+      { type: "word", sourceIndex: 7, sourceEndIndex: 13, value: "\\35 -0" },
     ],
   },
   {