I signed up for the $10 plan, before they introduced unlimited searches. One month I went over and thought ok, I'm not really ok with making more than 10. A few weeks later they switched to unlimited.
The lens feature is great. I have one for searching my company's public documentation etc.
The redirect thing is great too, it can rewrite URLs on the fly, e.g. Reddit -> old.reddit.
I have used other search engines in the past, like ddg for a long period, and others. I always ended up just doing !g on every search.
Kagi is the first where I don't notice I'm using another search engine and actually if I'm on Google I notice my experience is worse!
Great to see updates like this, in the months I've been using it it has improved so much already.
If it can automate for me navigating to Reddit, then modifying the url to old.Reddit, and hopefully not accidentally clicking on the page and needing to reverse back or get redirected based on the content, then that’s probably worth 10 dollars a month already
In general is that you can't ever trust them apart from a few very large ones.
A redirect extension is ripe for the usual blackmarket shenanigans.
I have my own omni extension made up of copy pasted open source extensions. I normally wouldn’t go through the effort but web browsing is crucial and frequent enough to be worth it and it took less than a weekend to get it up and running, especially with Plasmo. It’s also another reason to prefer Firefox and MV2
It's the auto-update behaviour of browsers that should probably be criticised here, rather than having general distrust of all extension authors.
But it's not just reddit, you can do like a custom regex replace on any domain IIRC.
You would think a mega company would think more long term then that
Kagi basically won me after I realised I can exclude whole domains from search results - now Pinterest, Stackshare and all those AI-generated content filler sites are gone from my search results. Search results are great and it's just worth it.
The only advantage Google so far has IMHO is location-aware search. For certain keywords (like "gym") it feels like Google better looks for results that are within your area, whereas Kagi needs some more input (like a city/district name).
edit: I just realized there should be a search bang for this and indeed there is. Adding !countrycode to your query works. Still a bit tedious on phone, but helps on computer.
Kagi has gotten better at directly providing answers (like 15lbs in kg) but it's still lagging.
The google whatever-box is more frequent than the one at Kagi. Is it better?
The number of times I have googled a question and had weird and probably completely false information inserted (using reddit as a source, no less) is ... well, it's more than I'd like. It's at the point that I try to deliberately ignore the whatever-box at the top of the google search results page.
And if I google for a product, the whatever-box gives me some very useful... price aggregates? ads? reviews? Who knows.
I certainly wouldn't go back to google search just because the whatever-box appears more often.
Edit: I just saw that Kagi supports "regional bangs", e.g. !nl to search results for The Netherlands. This is, for me, almost on par with DDG
Example where this annoys me with Google: if I search for "gdb", the first result is a local burger place named GDB, and not the GNU Debugger
Actually, you can configure this with Kagi and it works really well, at least on my area.
I have an extension that does this for me
I asked the developer for Google Hit-Hider to add the ability to block a 2ld (gov.br) when confronted with many results from 3ld (crap.gov.br, junk.gov.br, chum.gov.br) - and he did. You can trim the domain right in the action window. JS is one of the best people.
Sort of. You can block the domain in the result but not the spammy domain underneath.
If you're getting tons of result with gov.ar (spam1.gov.ar, spam2.gov.ar, spam3.gov.ar) you have to use up your blocklist for each one. You can't block gov.ar.
We can't edit our blocklist directly.
Google search feels more like shopping search nowadays.
"Drift to low performance is a gradual process. If the system state plunged quickly, there would be an agitated corrective process. But if it drifts down slowly enough to erase the memory of (or belief in) how much better things used to be, everyone is lulled into lower and lower expectations, lower effort, lower performance."
-- From the book "Thinking in Systems".
Perhaps better than Google, but the whole paradigm shifted
When I ask an AI tricky questions, the responses usually point me in the right direction without fully solving the problem, so those citation links are really useful.
With Bard in particular, I find it often gives a confident but incorrect answer. When I point this out, it apologises then gives another incorrect answer, and so on.
Kagi, if you're listening. Please.
- Payment can be made in multiple forms which are not tied to a customer identity (you can even mail them cash) - Accounts are not username/password, no personal info is requested. When you pay you're given a randomly generated account number (currently 16 digits) which is valid for the time paid and each connected device, up to 5 simultaneously, will receive a random name.
More info here: https://mullvad.net/en/blog/mullvads-account-numbers-get-lon...
This is a great suggestion for opening up as a more general purpose search engine. The history on/off option is nice but not an adequate guarantee for many situations. Searching for regionally banned or "unfavorable" topics while traveling or living under a regime, consulting legal information, or just plain old porn searches are all situations where one might consider having their name, email, and credit card stored with the potential to be linked a bit too risky.
There are only two reasons I still visit Google:
1. I'm based in the UK and often it shows me shops based
in the US high up within my search results.
2. Sometimes I specifically like to visit Google Shopping
to find the lowest prices online for an item I'm trying
to buy.
I just don’t use search enough to want to pay. Perhaps that is because it is not so great though.
I hope they will continue to focus on their core product and don't branch out to do something else.
I absolutely did not think I would pay for a search engine. Especially as I've recently been on a quest to cut every unnecessary subscription. YouTube Premium is gone (thanks to 83/% family plan price hike). Half of my dozen or so domains are gone... you get the idea. Who would pay for search, right!? I signed up for the free trial, fully expecting to walk away.
But have you noticed how the quality of Google has steadily deteriorated over the last decade or so? How it now just surfaces ad filled garbage, listicles, and regurgitated content, ad nauseam?
I'm sure it's not just me who prefers what Google used to show around 2011.
Kagi is like the Google of 2011. Except, better. You can combine a bunch of domains you think are related and create a "lens" to just search them. It comes with a bunch of pre-built lenses, so you can search for academic results, or forum posts or pdfs. You can tell Kagi if a website is good or not, and it will then tailor your results according to your preferences.
It kind of feels like what Google could have been, if the brilliant minds employed there weren't just tasked with chasing billions more in ad revenue every quarter.
Honestly, its results are better than Google's. Forget the lack of ads and them not tracking you and not selling your data (from which Google Search apparently makes on average $300 USD every year from every user). Just purely based on the search results, Kagi is better.
I've recently converted my plan to a "Duo" plan, and the SO, who doesn't over care about privacy and ads and stuff, also likes it. #win