A response can carry HTML while also declaring gzip or another content encoding. Those labels describe different layers: what the content is and how it has been encoded for transfer.
Keep the headers in separate columns
The MDN Content-Encoding reference explains how the header identifies encodings applied to a representation. Content-Type still describes its media type, such as text/html. A browser can decode the response before interpreting the HTML.
When inspecting a request, record both values. Do not conclude that a page has become a different document format simply because its transferred representation is compressed.
Compare the right sizes
A browser’s network panel may distinguish transferred size from decoded resource size. Read the panel’s labels before treating two displayed numbers as contradictory measurements of one thing.
For a practical check on your own site, inspect a small known page and identify the encoding actually used in its response. Avoid assuming every resource follows the same configuration. Understanding this layer helps explain why the bytes sent over the network can differ from the content the browser ultimately works with, while keeping the document’s purpose unchanged.