diff --git a/httpie/downloads.py b/httpie/downloads.py
index 9c4b895e6f..57e2b722fa 100644
--- a/httpie/downloads.py
+++ b/httpie/downloads.py
@@ -218,8 +218,17 @@ def start(
 
         # FIXME: some servers still might sent Content-Encoding: gzip
         # <https://github.com/httpie/cli/issues/423>
+        # Skip Content-Length validation when Content-Encoding is present.
+        # Per RFC 9110, Content-Length reflects the encoded (compressed) size,
+        # but the requests library transparently decompresses, making written
+        # bytes exceed Content-Length. Skip the check to match curl/browser behaviour.
+        # See: https://github.com/httpie/cli/issues/1642
         try:
-            total_size = int(final_response.headers['Content-Length'])
+            content_encoding = final_response.headers.get('Content-Encoding')
+            if content_encoding:
+                total_size = None  # cannot reliably compare compressed vs decompressed bytes
+            else:
+                total_size = int(final_response.headers['Content-Length'])
         except (KeyError, ValueError, TypeError):
             total_size = None
 
diff --git a/tests/test_downloads.py b/tests/test_downloads.py
index b646a0e6a5..61e090f9ec 100644
--- a/tests/test_downloads.py
+++ b/tests/test_downloads.py
@@ -12,6 +12,7 @@
     parse_content_range, filename_from_content_disposition, filename_from_url,
     get_unique_filename, ContentRangeError, Downloader, PARTIAL_CONTENT
 )
+from httpie.status import ExitStatus
 from .utils import http, MockEnvironment
 
 
@@ -259,3 +260,17 @@ def test_download_with_redirect_original_url_used_for_filename(self, httpbin):
                 assert os.listdir('.') == [expected_filename]
             finally:
                 os.chdir(orig_cwd)
+
+    def test_download_gzip_no_false_incomplete(self, httpbin):
+        # Regression test for https://github.com/httpie/cli/issues/1642
+        # Skip Content-Length validation when Content-Encoding is present.
+        # Per RFC 9110, Content-Length reflects the encoded (compressed) size,
+        # but the requests library transparently decompresses, making written
+        # bytes exceed Content-Length. Skip the check to match curl/browser behaviour.
+        r = http(
+            '--download',
+            httpbin + '/gzip',   # returns gzip-compressed JSON
+            env=MockEnvironment(),
+        )
+        assert 'Incomplete download' not in r.stderr
+        assert r.exit_status == ExitStatus.SUCCESS