I don't get this either. In the bottom screenshot. LCP, it says 3.8 sec, but above, it says 1.1 sec, and they're both from pagespeed.web
The first screenshot is a live test using the tool server,
The second one is reporting the actual loading speed for users browsing the website/page,
If it reports the loading speed for the last 90 days instead of after validation, then what's the point of hitting validation then
If you're referring to page speed insights (or Gtmertrix in your case), then it's just used to make sure your website is well-optimized for speed,
It doesn't guarantee fast loading speed for users, as mentioned before, the server location, users' internet connectivity, page(s) size, etc... contribute,
If your website is hosted on some server that's too far from the majority of users, they won't get it loaded as fast as a near one,
If most of the users are within 3rd tier countries, where most of the internet connection is trash, the website won't be loading as fast for them,
The same applies to page sizes, bad internet can't load a 2MB page as fast as the servers used by these tools,
As well as weak devices that could take more time to render these pages, advanced/complex JS codes & design use more device resources,
For your case, you can either #1 ignore these reports, it's not a critical problem to over-worry about, it's not critical,
Or #2, change the hosting location if the server is too far, minimize the pages (images, design, animation, JS, etc...), basically making the website minimalistic
