I think it uses the same functionality as the guest account feature in the Android app as of now.
My private Nitter instance is also working again after the latest set of changes: https://github.com/zedeus/nitter
"Nitter is a free and open source alternative Twitter front-end focused on privacy and performance. The source is available on GitHub at https://github.com/zedeus/nitter"
No need to sign-in to view Twitter.
To add, Twitter's backend has a bunch of layers on top of it. I did a bit of reverse engineering for my blog article on scraping Twitter [1] and there's a bit of magic token/header generation involved.
Good resource if you'd like to learn more about how the whole backend works is Nitter [2] which implements most graphql functions in Nim (it's very readable even if you don't know Nim).
1 - https://scrapfly.io/blog/how-to-scrape-twitter/#scraping-pro...
> Uses Twitter's unofficial API (no rate limits or developer account required)[0]
Not sure what that means
It has some nice features, like lenses, URL rewriting, and the ability to bump sites to the top.
I have a lens for programming. It shows only certain sites, like GitHub and Stack Exchange, and I enable it when appropriate.
For URL rewriting, I have Kagi set to rewrite URLs from Twitter and Reddit to my own self-hosted instances of Nitter [0] and Teddit [1], so that when I open a Kagi result that would've gone to twitter.com, it goes to nitter.mydomain.com, which lets me use Twitter without an account.
Lastly, I use the site bumping feature to move Wikipedia results to the top.
One public instance: https://nitter.kavin.rocks/
Github repository: https://github.com/zedeus/nitter