I’ve been using Kagi full time and I like it a lot. It’s been worth the price.
I expected to like lenses and favoring/blocking specific domains. What I didn’t expect was how much their “Quick Answer” would change how I search.
I’ve been “AI hesitant”, in general the chance that an LLM will hallucinate makes these kinds of tools more trouble than they’re worth for me personally. In Kagi’s case, though, the individual facts it states in the quick answers have citations linking to the site it drew that information from.
Here’s what I’ve found:
- it’s been accurate most of the time, but not 100% (as expected)
- citations are pretty accurate most of the time
- every so often the citation links to a page that seemingly doesn’t back the claim in the quick answer
Unsurprisingly, I don’t trust the AI generated quick answer in isolation, what it does do is let me scan a few paragraphs, find the one that answers my question most specifically, and visit the sites it links to as citations for that piece of the answer. This saves me the time of clicking through the top $N results and scanning each page to find the one that seems to answer my query most directly. It’s like a layer on top of the page rank.
I remember using Google the first time and being impressed how the top answers were so much more relevant than Yahoo, it was a huge time saver. Now I find myself wondering if the “quick answer” citations will prove to be a similar jump in accelerating my ability to find the right web page.
It also makes me wonder if their own page rank algorithm could incorporate the quick answer output as an input to a site’s rank? That would be an interesting experiment!
Same experience with quick answer. I originally thought of it as a gimmick, but it's really changed the way I search now. The summaries are good, and the citations are amazing. I (nearly) always click through to the cited sites to decide how much I trust them, but the vast majority of the time the summary is spot on and extremely helpful.
On a side note, as a software dev I was curious about how it would "answer" queries that weren't questions, so I've tried feeding it queries like "linux distro" or other things that are most certainly not questions. About half the time it does a pretty good job at answering! For the really hard ones, it seems to just morph it into "what is <query>?" so nothing real complicated, but I did find it interesting.
Kagi once returned nothing for one of my searches. I didn't anticipate that and decided to go to Bing. Bing returned many results but none of them was relevant. This is what any decent search engine should do -- return nothing, if you query is bad or too specific.
I used to get no results all the time, and it was very useful! Unfortunately, that seems to be happening less frequently for me. In verbatim mode with personalized results off, I noticed Kagi not respecting quotes for phrases. Google will ignore my search parameters intended to reduce results for free, so... :-/
If you have specific examples, I think kagi team would like to hear about, I would suggest that you open post them in support website [1] and I'm sure they will look into the details.
Yeah more and more Kagi seems to be trying to give me the same trash results Google did while ignoring parts of my search parameters.
I have never asked for or even wanted "personalized" results, because on Kagi and everywhere else, personalized is shorthand for "very very poor guesses". It's very frustrating.
Turning personalization off just turns off custom redirects and any website ranking adjustments you've made. That's really the only personalization we have for results.
Not if there literally is nothing that matches your query. There is a tendency for services to be scared of ever returning nothing, and instead they will return things that they think are related to your query but really aren't.
Example: If you search for a specific movie title on Netflix but they don't have it, then they will give you a list of movies that they think are similar to the one you searched for. That is because their database actually knows about the movie and therefore can find links to other vaguely related stuff, e.g. movies made by the same director, with a similar theme, etc. But if I search for a specific title, then none of this is what I want, and I don't want to spend the extra 10-20 seconds scrolling through the list to realize that they actually don't have what I want. This is clearly a search experience which is optimized for maximizing engagement rather than user experience because a small minority will end up watching something from the garbage results while the majority will waste their time and be burdened by extra cognitive load. Shareholders are happy, users suffer.
>Example: If you search for a specific movie title on Netflix but they don't have it, then they will give you a list of movies that they think are similar to the one you searched for.
I absolutely hated that when I was a subscriber. That 1/4 of seconds of believing the search will succeed, just to give me the subpar copycat of the movie I was looking for.
> But if I search for a specific title, then none of this is what I want, and I don't want to spend the extra 10-20 seconds scrolling through the list to realize that they actually don't have what I want.
This has never been the case. If it can’t find your title, it’ll display “titles similar to”, right at the head part of your search. No 20 seconds of confirmation needed.
I actually prefer Netflix’ way because if I search for “Demolition Man” and they don’t have it, it might be that I’m in the mood for any <2000s action schlok, and who says I already know about “Escape From New York”?
I just tried searching in browser and Netflix says nothing like "titles similar to".
I searched for "Ted Lasso".
It has grey text "More to explore:", white text "Ted Lasso" and then thumbnail list of different shows, and it's literally just thumbnails, you can't even Ctrl + F and you have to read all the titles in different colored and stylised fonts.
It's as if it is intentionally built in such a way to make it hard to understand that it's really not there.
And in TV it says nothing, just gives you the thumbnails and since it takes longer to type you must check after each character whether one of the thumbnails happens to be what you are searching for.
I'm not home rn so I can't test it, but I'm quite sure that netflix says something like "we don't have X, these titles are similar" or smth like that. Maybe I just have an old version in my TV idk.
I've seen both. Sometimes it says it doesn't have it, other times it just displays results like it does have it even though it doesn't. You might be onto something, it's probably the difference between the web interface and app interface (on various devices).
> There is a tendency for services to be scared of ever returning nothing, and instead they will return things that they think are related to your query but really aren't.
With Netflix I assume they use data from IMDb for finding similar movies.
But one platform having particularly surprising ability to find “similar” things is AliExpress.
On AliExpress if you search for a brand and model of something without saying what it is, AliExpress is still sometimes able to know what kind of thing you are looking for and show similar products from other brands. And I’ve been wondering how they do that.
Maybe AliExpress has a big database of products that they scrape from the internet and classify, even for brands and models that have never been on AliExpress.
Or they could be able to do it based on similar queries that people made in the past where someone for example included extra keywords about what they were looking for. Or those people first having searched for a brand name and model and then made subsequent searches for more generic descriptions of what they looked for.
Or sellers could be including names of brands and models for products that are similar in the description or other input fields for metadata for their listings.
No. It is the single most important reason why I pay for Kagi.
It seems to me "everyone" think it is always about privacy or features or something.
But the main thing that keeps me on Kagi is the results. They seem to have most relevant results and few irrelevant results and if I decide to be specific using doublequotes I get no irrelevant results wrt that word. (And if you find one it is a bug and will be dealt with.)
I have lost enough hours of my life clicking through Google or Bing results that maybe has something relevant to my search.
Edit: I have been beating this drum since matt_cutts was in Google and used to frequent HN and so I think it is relatively clear that Google does not care about the quality of the search results.
So many times I'll often search something on Kagi, get no results, and tag on the "!g" at the end to see what would happen. Of course, I get a ton of results that have nothing to do with what I was searching for. I love Kagi.
Kagi uses results from almost every search engine in the world plus its own results. If you can not find something on Kagi, it is likely you will not be able to find it anywhere.
I can not remember a time where going to another page gave me the result I was looking for. If it's not in the top 10 it's probably not the right query.
This never happened to me until I started iOS development. Everything is built on top of layers and layers all the way down to APIs prefixed with “NS” for NextSTEP. Obviously, first the modern APIs are surfaced, but sometimes you really are looking for something deep, so you go deep into search as well, eventually finding stuff written in 2010 and such
if bad results are returned, you still can reword your query to match it better. I would prefer to see related, or slightly related results instead of 0.
I won't, because I have to actually scroll through those results to realize that they couldn't find what I want. It's like asking for where the apples are and then being led to an aisle with bananas, melons and pears. I'd much rather just be told that they have no apples.
Returning random unrelated garbage does not mean you have a higher chance of finding what you're looking for, it just means you're going to waste time sifting through useless noise.
And that is time you don't spend on refining your query, so it makes you actually less effective at searching if you consider people do not have infinite time.
You have a higher chance of finding something. I think you actually have a lower chance of finding what you want if it returns irrelevant results, because then you have to spend time manually evaluate and decide that the results are irrelevant before making another query.
Then it should suggest a better one and then evaluate the query anyway.
> This is what any decent search engine should do -- return nothing
WHY?! That's the opposite of it's job!
I have an account with a username that is spelled very similarly to a real word. Google will suggest searching for the real word instead. If you do that though, you'll never find the username!
I'm tired of people saying the computer should not do what I tell it to. It's like children who won't even attempt a multiple choice test because they aren't 100% sure
> Then it should suggest a better one and then evaluate the query anyway.
No the hell not. It should do what I tell it to. For a search engine that is to show me what it has about the query I input. If that is nothing, that's what it should show. It should not show me entirely unrelated results, ads, or what it "think" I meant. Not its job.
> No the hell not. It should do what I tell it to.
That's what I said. "then evaluate the query anyway." I should have added "original" to that statement
> If that is nothing, that's what it should show. It should not show me entirely unrelated results, ads, or what it "think" I meant. Not its job.
I'm saying I don't want similarity cutoffs. Most FTS methods involve a similarity score, I'm saying I don't want only results with > 0.1 similarity. I want all of them that were returned.
I'm NOT saying it should somehow inject results that didn't originate from the original FTS query.
It's sillier if you imagine the query in SQL. How can the database fulfill both "all queries have at least one row" and also "your WHERE clauses are interpreted exactly?"
> Then it should suggest a better one and then evaluate the query anyway.
Google does this, and they suck at it, unless you just spelled a word wrong. Do a niche or very specific query, for which Google has no answer and it will, without fail, remove the most relevant keyword and give you a bunch of junk results.
Have you never copied and pasted an error code into Google and have it return zero or only 1 or 2 results?
It’s terrible but far better than getting 100’s of irrelevant results because Google decided two words out of 10 in your query were the only ones that matter.
I actually have had this happen and it's infuriating.
I've had queries of copy+pasted errors with zero results, but playing around with it a bit just to find a github result that was only like two words off.
> it’s terrible but far better than getting 100’s of irrelevant results because
Surely the similarity to the one on github would still have it ranked on the first page?
> This is what any decent search engine should do -- return nothing, if you query is bad or too specific.
Yes but there's a >0% chance that you'll click on a potentially sponsored link (or a non-sponsored link to a page that itself contains ads) when you instead see a bunch of unrelated results. It makes financial sense to show random results vs not showing anything.
I strongly disagree on this. If a search with no results costs them about the same amount of compute as one with results, then satisfying that requirement would give them a commercial incentive to lie to you about whether they have any good results for you, and to waste your time scrolling through bad results. Your time doing so is almost certainly worth more than what the search itself cost you.
Kagi now is not only a search engine, the Ultimate plan gives you code, chat and research assistants. For chat you can even choose gpt4 turbo, gpt4, gpt3.5, claude2 or mistral models! On top they also have a fast summarizer. I honestly don’t know a single service that packs that many features for the that price nowadays.
I hope they don't stray too far of course with other features. Been using Kagi continuously since August, and I've developed a pattern where I still jump to Google when I need to buy stuff, get latest news and find local (national, or EU) content.
It's great for technical lookup, and answers, but for general search usage it's not there yet. I'd love to give actionable feedback on how they could improve, but can't figure out exactly what's missing on that department. It could just as well be a small crawled index.
I switched to Kagi almost two years ago. I have an experience opposite to yours. I never ever use the google bang. I did in the beginning, when a query wouldn't give me results, only to get worse, more verbose, equally useless results from Google. Quickly learned that if Kagi can't answer a query, Google will fare no better (and will waste my time with junk).
I'll note that to get local news, I do have to switch the region selector from "International" to "Portugal". Kagi doesn't have Google's behaviour of using my IP location. Which is good. Getting international results from Google is a struggle.
Been using kagi for awhile, also in Portugal. Google shopping is the one thing that Kagi can't beat them at yet. If i'm looking to buy a product online not from amazon it's still the best option.
I don't know what you're searching for, but I'm also from EU and I'm able to use Kagi for everything, including content from my own country. Sometimes I might have to change the country or language filter, but I've never had a situation where something was available on Google but wasn't available on Kagi. Nowadays, I'm only using Google for some quick answers, like showing the score of a football match or the population of a country. Kagi doesn't seem to support quick answers for these queries yet, although I have seen quick answers for weather, calculations and shipping and flight tracking. So I'm sure more quick answers will be added eventually.
Kagi has just also integrated with Wolfram Alpha, so it is now much better at answering factual questions such as population of a country or timezone. Wofram Alpha has always been more accurate than Google on these questions.
Showing the score of a football match, and doing online shopping, are the two remaining use cases I have for google. My usage of Google has now gone down. 95%. Kagi is simply better.
Wolfram Alpha seems to work quite well from my small tests, even with recent events. However, I don't really like to read an answer that seems like it's a person talking, I would rather just see the data in a big size like on Google. That's just a nitpick though, it's definitely useful. Thanks for telling me.
One example is that if I try to lookup a specific product I'd like to buy, I'm going to find more results on Google. Also (with the country filter enabled, Romania in my case), I still get results from US, Canada and other countries (which in practice is a hassle with delivery), but even if that weren't the issue, the pages are not in Romanian either to have a reason to show up in my results.
I didn’t realize they supported full conversations with my choice of model. They need to advertise this better! I just upgraded to ultimate and can cancel my ChatGPT subscription. Love that I can give more support to Kagi _and save money_ VS my pro plan+ChatGPT.
Kagi assistant is technically still in beta, gathering feedback. We are not satisfied with the experince yet and are working on an overhaul, planning to officially launch it in March. This is why it is a bit 'hidden' from view, but yet available.
It depends on the feature/product. They have custom models for some things (and honestly FastGPT and their summarizer gives pretty good results), but when it says it uses GPT4, it really does.
> In the same spirit of getting answers faster, now simply starting your query with an interrogative word (what, where, who, which, when, how) or just ending it with a question mark (?) will automatically trigger Quick Answer:
This is the type of attention to detail that makes me a happy user. Killer feature for me, no more having to type !quick (and yes I set up quick bangs but always seem to forget).
I have been really pleased with Kagi, though I do live in perpetual fear that they'll be acquired. Please no. And God help us if Apple acquires them, which seems like a risk given Kagi's great work on search and AI, two things Apple really wants. Don't do it Kagi!
The founder has previously said that they're bootstrapped specifically so that they don't have to chase an exit to get a return for investors. Obviously that doesn't mean they won't change their minds at some point, but it does give a lot of comfort.
I also live in this fear and have specifically communicated it to Vlad in an email before. He gave the same response as one of your other responders - they have no intention of selling.
Of course, that's the only answer any successful founder would give you...so I still live in fear :(
Kagi founder here. I wake up every morning loving my job, our member community, our mission and the change we make in the world. We are basicallly sustainable at this point and I am in awe of the opportunities in front of us as a user-centric company, bringing a friendly version of the internet to homes worldwide. Things are looking good, we have no plans to change any of that.
Thank you for all of your work! I've been with Kagi since early in the beta and it's just been getting better and better!
Just this morning I realized that when I look up a classic Disney movie Kagi turns up the classic and not the 2010s remake, where Google invariably turns up the newer one. Little user-centric things like that are why I love Kagi so much!
About a month ago I noticed I could no longer use Kagi search from Firefox 78 esr on an aging OSX 10.9.5 laptop I like to use due to a Javascript problem. GitHub's directory view also seemed to stop working on it about the same time with the same tell-tale Javascript error.
The latest Firefox ESR I can run on my old laptop doesn't like ??= and some other newer JS syntax. I see this sort of problem in my own daily development and I know with tools like esbuild you can simply specify --target=firefox78 and end up JS that works fine on older yet perfectly working hardware.
So what does this have to do with Kagi? Well after a bit of looking around I found https://kagifeedback.org and reported the issue [1]. A day or so later Vlad triaged it and it was fixed within the next few days.
A month later the problem reappeared in another place[2], Vlad triaged it again and it was promptly fixed.
Needless to say I'm very impressed with attention Vlad and people at Kagi give to problems and the effort they spend improving their product.
Kagi is good, I’ve been an unlimited subscriber for a while. Two low hanging fruit opportunities though:
1. Location based or aware searching - it keeps giving me the UK versions of storefronts or the identically-named steakhouse a thousand miles away and not the one I can walk to. Kind of lame, particularly giving me random countries’ versions of sites.
2. The map — maybe this is a setting, but the map always gives me nearly useless results. This is probably tied to geolocation or location aware results.
One of my main complaints (aside how bad the iOS extension is, but I guess Apple can shoulder most of the blame there) is that I often get US storefronts despite using the "United Kingdom (GB)" region. How odd! I even get amazon.com results, but never amazon.co.uk.
The iOS extension has a kludgey install, but it works perfectly for me. Did the weird stuff once, never had to think about it again , and all I get are Kagi results.
I've had to disable it since a recent update as it broke the "!g" bang which I still occasionally find useful. It seems to insist on intercepting Google searches even when I explicitly use Google. It also ruins the UX of the address/search bar somewhat (my search term becomes a Kagi URL so is less easily editable), but that is Apple's fault.
I constantly get links to the UK version of sites when I search for things from the US. It’s very annoying, and sometimes sites are broken in a weird way where a cookie is set and I can’t get to the US version without being redirected back to UK version.
I agree that maps, looking up locations etc is the one ingrained Google-user pattern that always leads me to a weird dead end and has me popping back over to google for JUST that query.
I don't know if there's a toggle I just haven't found yet for just automatically using Google Maps for location based searches until their feature gets a little better.
Beyond that, Kagi has been an incredible breath of fresh air.
Forgive my ignorance, but how is this “low hanging fruit”?
Google is still extremely competitive in the maps/location space, and I’ve heard Apple Maps is very good these days. Secondly, these features are not trivial. Maps are much more interactive - giant teams or even orgs are managing these products at faang.
On the other hand, having language-awareness is perhaps more important. Google frequently gives me results based on location, although I don’t speak the language (well). Getting results in English is hard where I am, in the typical infantilizing fashion of megacorps (we think you want X, and we’re not gonna offer you a setting).
Edit: apparently Google added Spanish as a language automatically based on some searches, and then search results for say API docs came up in Spanish first instead, despite English being my “preferred language”.
Whether we like it and use it or not (I love it and use it), I think we can only be super impressed by the productivity of the Kagi team. Look at these dates. It's insane.
I’m actually currently pretty annoyed with them, as it’s been over 2 weeks of non-English searches being borderline-broken [0] with no feedback or ETA.
You aren't using the correct language though, you are on International. Local language searches has improved vastly lately.
Now I can just keep it on my local language, even while doing English searches. Previously I had to switch between national and international all the time, that was a pain.
Nope, my issue was the one further down with `!de`, it got merged into this one.
> Now I can just keep it on my local language, even while doing English searches.
That’s exactly the issue. I search for a name, so it can’t know what language it’s in. But I want the German results, that’s why I switched it to German.
> Previously I had to switch between national and international all the time, that was a pain.
That wasn’t an issue at all for me, I use !de for German searches, and nothing for automatic international searches.
Yeah, seems like we have totally opposite opinions, I vastly prefer the way it is now where I don't have to use bangs/change settings all the time to get good results by default.
My language, Norwegian, is also a much "smaller" one than German, so search results without any English are outright worse. Personally, even if I spoke German, I'd prefer to get Wikipedia on top for a search like "Olympus E-P2".
It seems like they should implement a "verbatim language" search option, so both use cases can be satisfied.
With kagi, can't you block and/or de-rank specific domains? This obviously isn't a fully general solution, but if there are a small number of domains that A) come up frequently and B)you never want the english version (like ebay and amazon), then just block the english versions of those websites and promote the german versions.
To be clear, I'm not saying that "this is already good enough so what are you complaining about", I'm merely suggesting something that might go most of the way to fixing it until Kagi eventually implements a better search. They are a small company and I'd imagine that non-english language (and especially multi-lingual) behavior is not going to be the top-most priority, as understandably frustrating as that is for the people who need it.
That's not useful at all. I normally want those pages, but not when I'm doing a German search.
Kagi is not broken in general, 95% of the time it's still fine because almost all my searches are in English, so without !de in the German region. It's only the regional search that is not working properly anymore because they went further into "we know what you actually want to search for" territory. Not doing that was a huge part of what I loved about kagi.
In addition, this worked fine all the time. I've been using Kagi since the earliest betas, they only broke it very recently.
It doesn't have a German Wikipedia article, so that's an impossibility. Agree with the store links though, except perhaps one US Amazon link - when I buy expensive tech I usually check 1-star reviews to see if it has serious problem.
I wouldn't be surprised it make huge difference. I would like to point to [1] compmike's comment. You get 80% Ruby along with Types benefit plus many Times the RUNTIME speed improvement.
I switched full time from Google to Kagi last month and it's been an improvement, or at worst a neutral change. There was a frustrating period of adjustment as I had to learn what Kagi's limitations and quirks were. Google has them too but I've long since learned how to compensate for them. Now I know Kagi's.
These integrations with other services and specific data are remarkably useful. Glad to see more! One weak spot of Kagi for me is their local / maps search. It's Apple Maps which is still not as good as Google Maps. But moreso Google Local has become a very strong product with a unique database. Nothing stopping me from using Google for local business info when I need it though.
I wish services like Kagi and others would make it easier to contribute to OSM so more people would update places information (like business names, opening hours, etc).
It doesn't help to rely on Apple Maps when everyone can contribute to OSM directly. It's also easy to make it accessible enough (but not to much to avoid trolls) just to increase OSM contributors base.
That's what Kagi does with their Wikipedia oneboxes and I regularly open the article to correct it when I notice an issue. I wish Kago (and others) would do that with places, so we can avoid opening the Maps tab to have quick place info.
Disagree. It is up to the businesses to send their up to date information to map providers, because it is in their direct interest that customers find them. If they're too lazy and cheap to do that, then it is their problem that they get fewer customers. They shouldn't rely on free labour and you shouldn't do this for free for them. If they don't know how to put their info on a map, then they should hire an agency that does.
Imagine being a business owner and ignoring the very basics...
That's it. And still they don't bother to take care of their properties on Google Maps. Most won't even bother to submit their stuff to Apple Maps.
The thing is: Submitting your info costs nothing in money and almost none of your time, but brings enormous benefits in bringing in customers. So if the business owner is too lazy to do it, why should anybody else do it?
If you submit correct info for your favourite local business to a map service, that might bring them thousands of new customers and the revenue from them. Then go to them and tell them you'd like a coffee since you put their info neatly into the maps. They'd probably spit in your face. Then turn around and pay thousands for Google Ads or for a local social media influencer to bring in 50 new customers.
Business owners being dumb about their internet presence is why for example Booking.com is a hundred billion dollar company.
Kagi has been my go to for some time now. For the past few months I rarely consult others.
I’d say toward the end of 2023 we passed the inflection point where Google is now generally worse. In fact these days Google is often worse or at least no better than Bing and DuckDuckGo. Kagi is better than all of them.
It seems like the Internet being winner take all is only part of the story. The reality is winner takes all, then turns to shit, then an unbundling occurs.
Nice! It works pretty well and the subscription lists are a great feature. Thanks for suggesting it. I was able to import all the block rules from Kagi there. At least for blocking it's great but I also lower/raise results in Kagi so that's a bit harder for this extension.
Still like Kagi better for the results (especially technical stuff), but I do still use Google for regional/news/shopping searches or things where Google Search behaves more like an assistant.
I think this is an area where Kagi will have some trouble competing because of Google's data moat and the fact they have been honing non-web integrations for a long time. This long tail must be 90% of the job.
As an aside, and I said this in a thread yesterday but, what I'm hearing about Kagi (I've not used it yet but I'm signing up today to try it out) and the contrast between the Sora and Gemini 1.5 announcements make me genuinely concerned about the future of Google.
Search is so much their golden goose and they've let it languish and now there are real dangers to that product from several flanks. They have Gmail after that but is that enough to keep the coffers full?
I had so much hope in Google Workspace as competition to Microsoft 365 but even that is in trouble if you ask me. I've been helping a company migrate from Workspace to M365 and one of their reasons was because they couldn't get any support from Google. Also, Workspace is hard to use. Not that M365 is easy, or that their support is any good but, Google has clear perception problems here.
I almost never hear of anyone talking about GCP anymore so I'm guessing they're losing mindshare in addition to market share.
Dart and Flutter are wonderful projects but it seems to me that since there's no way to stick ads on them, Google is content to give that department a little money and let them do their thing. But, of course, when some of the aforementioned products start losing more money, great products like Dart and Flutter will certainly be cut.
Maybe I'm looking at it the wrong way. Maybe Google's time has come and companies like Kagi can come in and fill the void Google seems hell-bent on leaving behind.
“Even though Microsoft (with 23% of the market) and Google (with 10%) each gained just one percentage point of market share year-on-year, that’s more than AWS can claim, which continued to hover around 32% market share.”. https://www.sdxcentral.com/articles/news/microsoft-and-googl...
Summarizing what I read in the threads the other day,
Gemini 1.5 was "announced" and you can't use it or see what it really does, it's just "better". (Admittedly, the context length sounds incredible, but we can't really test it and hype only works when there are no alternatives.)
Sora then dropped immediately afterward and was showing numerous really impressive demos and then taking user requests all day.
I recently stopped using Kagi after using them for a long time. For whatever reason, their results became pretty bad compared to before, often almost mirroring Google.
I am still a user because Google is worse than ever, but I wish Kagi would stop interpreting my queries and randomly decide to ignore or assign a random weights to my words.
If I search for `foo bar baz`, I want pages with `foo`, `bar` and `baz`, please. No, verbatim doesn't cut it because it somehow is far too strict and even common queries return half a dozen results at most. Google started sucking in the 2010s when they decided they know better than me, and the more Kagi improves, the more they follow down Google's path of trying to be outsmart me, and falling short.
I just want a 2003 Google experience. A computer is not smarter than me, stop trying to interpret and make assumptions.
---
Aside, is it so hard to make a search engine with a query interface like any trigram-based SQL full-text engine? With quotes, AND, OR and -word. I still believe any AI-powered solution of the past 20 years to be a mockery of this gold standard of precision and simplicity, except maybe Google 1.0.
That is only going to get worse, with the direct copies of StackOverflow and other popular sites (which are easy enough for search engines to filter out because the body content is practically identical) that exist only to fool less careful clickers into getting their adverts & such, being joined by a multitude of LLM generated rewordings of the same which will be harder to automatically drop (or de-prioritise).
Did you start to notice things getting worse around the new year when they added Brave to their search results? That was the turning point for me, I was a huge advocate but then quality went downhill and then a few weeks later they proudly announced that they had spent a third of their funding on...t-shirts. Between that and this huge push towards being just like Google and Bing with LLM hallucinated nonsense, I'm not sure what their aim is anymore.
Kagi is partly powered by Google, so I can imagine specific searches resulting in results pretty similar to Googles, if none of their other source have any matches.
I noticed something similar with Ecosia, they apparently switched to using both Bing and Google, and the results, while still pretty good seems a little worse. That's all anecdotal and may be completely random, but I think one issue could be that Google is getting worse.
I tried a few searches side-by-side and they were pretty much identical. I get that Kagi offers customization etc but that’s not worth the money for me personally. I just went back to Google because the results were the same.
Damnit I was on the narrow path of convincing myself to start paying for Kagi. I've been using QWANT but the search results are a bit off tbh. And they've started doing full-screen ads to turn of my adblock (which well to be fair thats the way they earn their keep).
For what it's worth, whenever I've switched to DuckDuckGo, I've found myself adding a !g to nearly every search. Since I've switched to Kagi a few months ago, I've hardly ever had to go to another search engine. If you can spare it, I'd suggest you invest the 5$ for a small account and just test it out for a month.
For me, it’s not only the price you pay. Having a username attached to each your searches is the opposite of privacy, no matter how much you trust their legalese wording
I go the opposite way. I trust a company that takes my money to pay its costs to keep my privacy. As opposed to a company who "doesn't know who I am". (Apart from unique fingerprint https://amiunique.org/ over many queries over many months)
They try to. But every time a browser gets better at existing methods, it also gains new features which become new methods of identification. The web browser environment is broken. And if you get to fight it by using something with better protection, you're only exposing yourself more because no other users do the same thing.
99% of people are probably signed into a google account when they search google. If you want the sort of privacy you seem to want, stick to the dark web and similarly inconvenient tools.
To maintain your level of privacy requires a drastically different lifestyle than most of us have or want.
They didn't state anything about their desired level of privacy other than not wanting a username attached to every search. That's not equivalent to needing to use TOR for every search.
>99% of people are probably signed into a google account when they search google
What is the relevance of this statement? Obviously this wouldn't apply to the parent poster, it doesn't refute anything the parent said, and at least anecdotally it's not true at all.
>If you want the sort of privacy you seem to want, stick to the dark web and similarly inconvenient tools.
Privacy is a spectrum, not some binary choice between having a username (and often real name) attached to every search vs. using the dark web for every search. You can land somewhere in the middle, for example: "I don't want a username attached to every search".
There is no technical reason Kagi needs usernames, but they choose to require them. For some people, that points to the company not being as privacy-friendly as other people seem to think/claim.
I dont browse signed in to my Google account, which I barely use. But Google has an open account on each of us, even if we block their servers at the firewall level
Oh well I don't assume I'm in anyway "anonymous" no matter if I got a username/ipadress/cookie/deep-state-monitoring attached to my searches. Things I html-POST to the web is no longer private.
Check it out for your use case. (Have you used up the free searches?) I've not noticed any degradation for my searches. It's going to be different for everyone.
I used them when they first announced their service. Was a nice experience afaik. QWANT is more like internet on hard-mode.. They don't really give you what you're looking for but close enough.. But I've become more dissatisfied with the results of late.
I'll mull the decision for a while but I'm probably going to give Kagi another try.
I noticed that search queries required more thought and perusal to get desired results than normal. Kagi previously would give good enough results that I could figure it out without having to think about Kagi itself. More recently I noticed that I wasn’t able to find answers like this anymore, and compared the query to Google. The results were pretty much identical. I kept using Kagi for another month but it persisted so I stopped my sub. I was an early adopted with Kagi and initially was very happy.
Since Kagi became Google for my purposes, I started using Google again. I don’t have much motivation to curate the customization features personally.
I would imagine that it could depend on the general type of query. As a developer I definitely noticed a decline in quality.
I’m quite privacy focused but for whatever reason don’t care if Google knows that I want to know how to write a helm chart or eat pasta later
I'm still using & paying for Kagi, but I have noticed a sudden drop in search results quality in the last month or two. Unfortunately I didn't take note of the specific searches, but I jumped over to Brave Search and everything I was looking for was right on the top, instead of down around page 2 of Kagi.
There's more blatantly obvious spam results seeping into Kagi as well, though at least I can block those domains when I come across them. A few of them were hacked websites, notably government websites in South Asia that had been thoroughly compromised.
Nope, I don't think so? I'm going through my settings now, and I have Safe Search and Image Safe Search both enabled. Which makes it even more surprising that it was returning hacked government sites mentioning porn in the search snippet.
(Just found what you meant by Verbatim search - nope, I would have always had Personalized turned on instead. Seeing those results would be the first time I even encountered those domains, so I didn't know to block the domain until then. Since the domain is still in my blocklist, I can confirm it was a government website in Mumbai. I don't have any reason to be visiting Indian government sites.)
I'm not saying every Kagi search is infected with hacked results, it's only common at times when I'm really researching something and need to get to a second page of results.
I also note that some Warez sites rank highly in Kagi too. Knowing that Kagi's customer base is largely from HN, that's less surprising to me, and I'm thankful that I can block those domains from my own personal results.
I haven’t had an issue. I recently started getting GPT results, in addition to the normal search results, and they have been very helpful for quick answers to questions.
Great to see continued improvements. One area where I continue to use google is specific local information, like opening hours, commute timetables. Try these two searches in Google and Kagi to compare:
"train from Galway to Dublin"
"Ilac shopping centre opening hours"
Kagi gives you a bunch of links that might be absolutely trash, if the service provider (eg: rail company) has a shitty website. Google gives you precisely what you're looking for, eg: a table of departure/arrival times for the next 5 trains, with a date picker. It can even handle the atrocious TFI (transport for ireland) car-crash, providing me with the precise info I need to get a bus from one part of Dublin to another. You can't even get that with the official TFI app, and they notoriously block access to their data API (breaking the only decent Play Store app for buses in the city)
I'm not sure if this is because Google have a budget for curated content like this, or if they have some magic way to extract that data without making a balls of it.
> or if they have some magic way to extract that data without making a balls of it.
They have a system where public transport operators can upload their routes in a specific and quite complicated file format. Google does not integrate any public transit unless the authorised organisation uploads the data to them.
What do I need from a search engine? Not an LLM chatbot, not any of the auxiliary “smart” stuff. I want customizable results, and to be able to jump into searching quickly and intuitively.
(Decent results, sure, but that is subjective.)
So, a big reason I was using DDG over Google, despite often subpar results, was bangs. Now I pay for Kagi, and guess what convinced me to subscribe? Bangs indicate caring about power users. How cool would it be if I could set Kagi as default search in any browser and get powerful shortcuts at my fingertips across all devices?
Well, for some reason Kagi seems to expect the bang to be added at the end, separated by a space, with exclamation mark after (not before). If you don’t do it just right, you’ll just be on Kagi, losing 5+ seconds and your train of thought.
I’m sure I’m in a minority but I swear, if Kagi is not going to make bangs work properly soon, I’m not renewing. If you do emulate DDG, go all the way, otherwise it’s just taunting your users…
As another datapoint, bangs through Kagi have been working perfectly for me on every device in the way you would expect (set it as default, and use bang as normal).
> Well, for some reason Kagi seems to expect the bang to be added at the end, separated by a space, with exclamation mark after (not before). If you don’t do it just right, you’ll just be on Kagi, losing 5+ seconds and your train of thought.
What?.. bangs work both at the beginning and end of the query, and the exclamation mark has to be put before the bang.
The number of times I landed on Kagi after typing !hn something, then tried multiple times to get the bang right, then went to Algolia’s instance direct, is maybe a dozen by now.
However, I wonder if they [ninja-]fixed it, because it does work as of now. Kudos if so!
From a programmatic and practical point of view, I don't understand why would someone want to put a bang in the middle of a request or within a word.
Basically you have your request and you add a criteria on it, not in it.
It's like ordering at a restaurant : you wouldn't say "I'll have a burger to take away, some fries and a coke". You either specify how you'll have your order before or after the actual order. And you don't create a new word like "a cokedinein".
Furthermore, I want Kagi to be able to reply to the request "how do the !g bang works?"; if I want something else to reply, I'll add the relevant bang around my request.
Oh yes because of the CSP. The CSP that allows forms that can change your settings... you could easily use the above bug to get some impact with an additional click on a form's submit button.
Admittedly, no full XSS anymore, but still dangerous and shows their lack of understanding and caring about security.
It's not the only place you can inject HTML and not every page has a CSP...
I don't get why they allow injection of irrelevant url parameters in the first place, it's the first rule of any input - remove what's not used and sanitize what is.
Regarding privacy: an obvious point is that you need to log in to use the search engine, so each search is tied to a unique user. Given that payment is involved, each user can be tied to a real-world identity.
So which search engine do you recommend that takes privacy seriously and that actually works (i.e. doesn't block me and return correct results) in Norway?
Marginalia Search's still a bit hit and miss. Haven't really actively been working on the search result quality lately, been too busy with various chores away from the actual search end of the search engine. It's gotten a bit better, but mostly by accident, through fixing bugs that had knock-on effects on quality.
A directed effort toward unfucking query understanding and index execution is up next on the list of tasks to tackle though. Hopefully it'll make a decent impact.
Like Google? Whose entire business model is pilfering that data without asking you? And who also asks you constantly to attach a phone number to your account, etc.?
Do you mean Kagi search function or their LLM? If it is the former, I use it as a daily driver and I do a lot of scholarly work and I rarely felt the need to switch to something else. Maybe sometimes I go to google scholar itself (not google search) but this again wouldn't be different from using google search itself.
I wish the maps functionality was better. When I want to find directions to a place, or see how long it will take me, the easiest thing to do it "cmd+L cmd+V" or "highlight --> right-click --> search for ____", but with kagi set as my default search, the UX is terrible. The top results are often for zillow or similar, instead of a map, and when it shows a map, it's many clicks and slow loads to get directions.
I’ve been using Kagi for a while. Unfortunately, its website on mobile is much lower quality than say DDG. One issue I find often is that when I want to edit a previously-entered query in the search box, sliding the text cursor is very hard – it often stops at search box boundary and doesn’t scroll the text inside. Another is that sometimes the search results page freezes after I come back to it from a different tab. When I try to scroll, it refuses to scroll and only briefly shows some white space around the currently-viewed part. I blame all those on some JS that’s too creative for its own good.
Aside from that, I wish Safari had a setting to add custom search URLs, so that I can paste my Kagi auth token, instead of having to use xxSearch, which also has some annoyances. Obviously, this part is not Kagi’s fault.
I’ve stopped using Kagi on mobile and went back to DDG. It’s still stellar on desktop though.
It is totally their fault. They should release a Kagi App for iOS and be done with it. No normal person is going to mess with hacky extensions and permissions to pay for a search engine. If I can tell people to download the Kagi App and search from there, all of my friends and acquaintances would be able to use it and probably happy to pay.
They actually have the Orion Browser - though personally I found it a bit rough on the edges so I'm fine with stock Safari + the Kagi extension.
I do agree that they should just release a Kagi app - or even better, just rename Orion to "Kagi browser" and have Kagi as a first-class citizen (with account login).
I've switched from Firefox to Orion on my work Macbook and I agree it's a bit rough. But I've never used Safari as a daily driver so I can't tell if it's Safari's fault or Orion's fault. The biggest issue for me is the terribly raw bookmarking UX (cannot drag-and-drop from URL bar to bookmarks bar, cannot move items around within the bookmarks bar, etc.)
They’ve got a new “blessed” method, which involves installing their custom Safari extension, which redirects. It works fairly well. Obviously not as good as Apple just not having a whitelist of search providers.
Google has local business hours because businesses provide them with this information. Kagi needs to purchase this data from Google if they need it, because businesses will only play ball with Google and in rare cases with Apple.
As for sports schedules and scores, I've seen this request a lot, but don't you have any dedicated website that you can bookmark for that?
I just do the google !g bang or !espn for sports. That said, if you look up an individual team Kagi will usually show their last games and results on the right
It's really cool to see a search engine alternative. I wonder about the practicality in the wake of large language models? It would seem almost easier to build a model that acts as a search engine than to build a traditional search engine.
Yahoo returns thousands of results for dorking queries where Kagi and Google return an identical list of 5-6 results. Pretty disappointing for me as a paying customer to learn that Kagi is basically a Google wrapper
I really need to try Kagi out this weekend. Whenever I try to change search engines I hit the same wall in that ~20 years of Google has trained me to think in terms of their search algorithm when searching.
Kagi is good, but they claim they're faster than others, but it depends. In my country, Cyprus (I know nobody cares), it's significantly slower than Google, so I decided to stick with a classic.
They have deployed to multiple datacenters, your nearest one should be eu-west. You can see the datacenter where you connect to by opening the drawer on the right by clicking on the settings icon in the top right, and at the bottom is the datacenter location + the latency to it.
I expected to like lenses and favoring/blocking specific domains. What I didn’t expect was how much their “Quick Answer” would change how I search.
I’ve been “AI hesitant”, in general the chance that an LLM will hallucinate makes these kinds of tools more trouble than they’re worth for me personally. In Kagi’s case, though, the individual facts it states in the quick answers have citations linking to the site it drew that information from.
Here’s what I’ve found:
- it’s been accurate most of the time, but not 100% (as expected)
- citations are pretty accurate most of the time
- every so often the citation links to a page that seemingly doesn’t back the claim in the quick answer
Unsurprisingly, I don’t trust the AI generated quick answer in isolation, what it does do is let me scan a few paragraphs, find the one that answers my question most specifically, and visit the sites it links to as citations for that piece of the answer. This saves me the time of clicking through the top $N results and scanning each page to find the one that seems to answer my query most directly. It’s like a layer on top of the page rank.
I remember using Google the first time and being impressed how the top answers were so much more relevant than Yahoo, it was a huge time saver. Now I find myself wondering if the “quick answer” citations will prove to be a similar jump in accelerating my ability to find the right web page.
It also makes me wonder if their own page rank algorithm could incorporate the quick answer output as an input to a site’s rank? That would be an interesting experiment!