I've been using Kagi for a while (almost two years now!) and it's been nothing but excellent!
Lenses are very useful (Reddit lens is on every second search), and I personally really like the AI features they are working on.
The new more advanced assistant which is able to do searches, which can also be constrained to lenses, and lets you pick an arbitrary model, is excellent, and basically means I don't need a chatgpt/claude subscription, as Kagi covers it very well.
All in all, great product which I'm happy to pay for.
I really wanted to like Kagi, I'm onboard with paying for search, but I've had had a big issue with its speed when doing the trial to be honest, am I the only one bothered by this?
Perhaps this is because I'm in Europe and it's faster in the US? A search request to Kagi seems to take around two seconds for me (shows as ~1s in the Kagi UI), it just feels really unpleasant compared to Google, I'm used to firing of a couple searches with different wording / terms and go through results quickly, feels like I'm being held back.
Maybe I'm spoiled, but if I'm paying for search I would really like it to be at least on par with Google, the search result quality seems ok from what I can tell, lenses don't really make sense to me, they seem to filter out too many results I would have liked to actually see, but the customization like adjusting the rating of individual websites is fantastic.
If they can manage to bring the speed to match that of Google, I'd be happy to pay for it I think.
I feel like search is probably one of the most common things I do with a computer, I would say I often search many times per minute if I’m actively looking into some topic or issue.
Because it’s such a common action for me, it feels like such a strong regression to go 2/3 times slower than before.
Back in the day, IBM did a usability study. An application requiring the user to specify operations and fields for data input was setup two ways:
One way was manual, lots of clicks, and or input sequences. Each one did a specific thing quickly.
The other way was highly automated and the user was only required to click a few times. The tasks were the same. This way had more flow, fewer discrete commands, more functionality woven together.
To their surprise the users felt they were more productive with the software they clicked more, despite the automated version being less work and the workflow more efficient.
I bet the effective reduction in your search is minor, but the delays do accumulate and demand attention.
Given that, a small change to your flow may well change things!
What you should do is rapidly input your first queries and then as they appear, drill down on those, and when that appears, start to eliminate dead query windows, or drop fresh queries into them.
What you prefer, to use a car analogy, is one that corners like no other. Then you find yourself in one that lacks corner cases.
So you maximize your time in lane, straight road, batching the corner driving and flooring it on the straights.
I used to experience a similar thing running a browser on IRIX vd NT. The NT browser was quicker to respond where the IRIX one would delay a little and then just render it all quick
I just started working with a few windows. Changed everything. I would be typing in new queries while one I waited for was about to render.
It was a change from rapid fire to a more batch mode. Soon, I rarely had to think and my flow was fast all around.
I put this shared experience here in the hope you may be inclined to try different things.
But do you seriously think you should get serious results in one second? I would understand the complaints about slowness if searches were taking like ten seconds, but a few seconds, I really must be getting old.
I still remember when google was giving relevant results in page 2. Now it's pretty much useless for me, and the fast search makes me think they are throwing away tons of potentially good stuff just to make it fast (and place more ads and rubbish scam/ai sites).
I haven't found that, but even if that is true, it saves you tons of time having to filter out all the sponsored results and get to the things that actually give you the information you where searching for.
That's not my experience, but I guess that depends on the use case and how specific things you are searching. However, in general I much prefer good search results to saving a few seconds and getting delivered absolute rubbish like google does these days, that was the main point of my post. Like I said I still remember when just plain google search without any thought on how to word the search was phenomenal.
Google fully pivoted from prestigious tech-company to lame ad-company and that's a shame really. Business as usual.
How bizarre. I'm not in the US either - I'm in New Zealand, and have been using Kagi since their beta I think and currently pay for Ultimate, and to me it's a lot faster than Google.
The other day I was using someone else's computer and used Google, and my goodness, the results were just awful and ... bloated?
Sorry to hear your experience hasn’t been great. I’ve been using it from NZ for the last six months and haven’t noticed any speed differences. Just did trips to Canada and the US recently and I didn’t notice any difference in performance.
Same for me in Poland right now - it takes around a second more for Kagi to produce the result than Google (with the delay happening after the interface of results page is shown). Doesn't bother me either.
What does bother me is the occasional failure to load anything at all, solvable with F5 key. But, being able to rank pages, or rewrite reddit.com to old.reddit.com more than compensates for this.
I am also based out of Germany, and experience the lag too. I find it very annoying actually, to the point where I am contemplating switching. The lag isn't always there though...
Possible reasons Google is faster for you is because
1. they simple pull ads from database on the first page instead of actually searching for what you need
2. They load marginally relevant answers instead of doing a better search for what you need
3. Google is multibillion company and can afford faster servers
With Google you waste a lot more than 2 second scrolling down past the ads trying to find answers, assuming you can even find it, and doing another search(es) if you don’t, wasting even more time.
Well, there never was any law that required robots.txt to be honored. Big players like Google do, but I am not aware of any consequences if they wouldnt (of course UNTIL it is then regulated).
I don't think there's anything special; I very often get new-reddit results with no 'site:' qualifier. I can't think of a specific query offhand, but it happens multiple times daily for me.
That disregards the fact that facts change over time.
If you google “best DSLR camera reddit” it’s much less valuable if the results are 5 years old, even if they’re LLM free - the cameras on the market may very well have changed in that timeframe.
I've been using Kagi for a while too (wow, 2.5 years, didn't realize), and agree that the experience and search results have been excellent. I don't use lenses, but make heavy use of bangs, and I like that I can up- and down-weight domains and even ban domains outright.
But I don't really care about AI assistants. If these AI integrations will improve the bog-standard search experience (I type something in the search bar, get a list of results, and click the one I think will give me what I want), that's great. But if not, this is just noise to me.
I was on the $10 a month plan for a while, but I canceled. Overall the experience of using Kagi was great but not "$10 per month" great compared to free alternatives. On top of that I got the feeling that Kagi is not a super "professional" company; for example by spending huge amounts of money on self-printing useless t-shirts (https://blog.kagi.com/celebrating-20k) or haphazardly suddenly charging tax on subscriptions.
What I want most of all from a search engine is to be "internet plumbing" and mainly stay out of my way, and with Kagi I always had the feeling they would suddenly remove/change/add things because of some strong opinion their founder holds. In which case I don't want to be $108 committed for a year.
I love Kagi and I’m a paid user, but I’m not willing to pay $25 per month for the assistants for the following reasons:
* I already pay these companies directly and wouldn’t be able to cancel these as I use the voice assistant on my phone from ChatGPT and love using the artifacts from Claude on my computer
* I’m also paying raycast to access these at the touch of my keyboard and prefer to quick access use it there
I love Kagi and can’t recommend it enough. I wish I could just give them my api key for this instead of paying several different service providers for the same ai access to the same models. This is getting expensive.
Highly unlikely, that would undermine their business model. For the same reason they don't offer pay as you go plan. Vlad explicitly stated somewhere that they are able to make money only because many people do not use their plans to the fullest.
There will certainly be a level where this is viable, if not sensible. Rather than having “casual” users cover the cost of “extreme” users, letting them specify an API key will likely be beneficial (albeit for a small number of users).
I mean folks would still be paying for an account, just not at the ultimate level. And since they offload the actual LLM processing to a third party, not much overhead cost there.
IMHO if you pay for them directly already, then probably not worth it. I cancelled Ultimate because it just wasn't feature competitive to me over OpenWebUI, but if you want to be able to try out several different models from different companies without giving each of them a card and using a different interface, Kagi Assistant could be a good solution.
There’s still a bit nuance to that - in most cases I’ve experienced integrated via API in a third-party app, the results have been mostly underwhelming, as opposed to being used directly, preferably with the configurable contexts. Including GitHub and Microsoft Copilot, various choose-your-AI apps, even corporate chatbots, sentiment analyzers and summarizers I’ve worked with. Asking via ChatGPT or Claude directly has produced more acceptable results to me than via an intermediary.
I’d wager this is something more to do with the system prompts when using ChatGPT or Claude from their respective app. Claude system prompts are publicly available. You “should” get the same quality from their API if you set the same system prompt they use for their app. I haven’t tested this but I think this is the right approach if you want to achieve parity between the two.
I’m not sure I read this right, but I think this feature is headed for the $10/mo plan, and currently exclusive to the $25/mo plan only as part of that plan’s early access to new features.
To balance the discussion a bit, I'm someone who pays $25 a month to Kagi instead of paying these companies directly. I like the easy access to the different models and being able to just google with Kagi "!chat (question here)"
For what it's worth, there's already a free Raycast extension that lets you supply your own OpenAI key which lets you do this pretty easily. Wherever possible, I lean towards the BYOK (bring your own key approach).
As a previous Alfred user, I would say that I miss the way that alfred searches files better, but everything else like plugins and overall usefulness find Raycast to be way way better.
I actually used this before but Raycast has some really nice presets or macros. I’m not sure what you’d call them, but being able to quicklink settings for different ai engines is really great.
I love being able to uprank, downrank, pin or blacklist specific domains/sites for my personal results. That alone makes it worth it. I also find the search results to be as good or better than Google. Once my personal ranking kicks in it's not even close.
Being able to personalize your own search is truly the killer feature, though, in a couple of ways.
The first is as you point out: being able to "edit your own algorithm" is really nice. I don't have to try to "train" Google's algorithm to show the results I want, and it's very easy to say "I never want to see this site in my results again". I'm still shocked Google doesn't have that feature even as some kind of client-side Javascript.
The second is Lenses. It's so obvious in hindsight that a singular algorithm is insufficient for search. Nobody wants or needs their searches for porn to impact their searches for technical documentation, or vice versa. There are more nuanced examples, but that's the most obvious (also, I don't think Kagi indexes NSFW content or at least I haven't seen any).
I've had this feature for years by using the uBlacklist Firefox extension. With a single click after a search, you can permanently block a domain from coming up in your Google search results.
And then gives you the astroturfed forum responses from “digital marketers” present in just-big-enough-to-matter communities. The rot is inescapable, just a different flavour.
>And unlike DDG, the quality of the results is actually great.
Any chance you can expand on this?
I've used DDG for years and find the results great, but maybe we just query for different things. Is there any specific topics that you find DDG is bad at but Kagi is good at?
I'm a paying Kagi user. I don't think that I can give a clear example or explanation of why they are better, partly because it's been a long time since I used DDG, and I gave up pretty quickly. The one thing I can say is that, for the short time I was using DDG, I found myself doing a search, and then in most cases, immediately redoing the search using the !g, because the DDG results were bad, and even though I am also dissatisfied with google results these days, they were better. Kagi has the same functionality, but I find myself almost never using it.
That being said, if you personally are happy with your DDG results, then I'd say you should probably stick with it. Kagi _might_ be better for you than DDG, but if you aren't actively dissatisfied (the way I was with Google and everything else), then the room for improvement is smaller. Might be worth using their free tier (which is 100 queries per month I think) just to test out some side by side searches though.
I personally switched away from DDG when they started removing results that were errantly removed by Bing. Since Bing is their upstream results provider, I didn't feel like I was getting away from the overreach of big tech, so I went to Kagi where they don't solely rely on upstream providers as they have their own crawler and such. The fact that they offer other great features like filtering results, and integrated AI makes it worth switching and paying for. Plus, I hate ads. Targeted or not. So I'm willing to pay to make that go away for a quality service.
I tried using DDG many times in the past. I found myself using the !g constantly, because I wasn't finding what I needed in the DDG results. After a while, I started preemptively adding !g, because I assumed DDG would fail me, as it often did.
With Kagi, I haven't had that. It finds what I need and surfaces things that Google didn't. On the one or two occasions I've had trouble and jumped to Google out of desperation, I also didn't find what I was looking for there... so I guess an answer didn't exist.
For sites I don't like, I can block them from my results, or rank things higher/lower. I can make my own custom !bangs and shortcuts, instead of being stuck with what I'm given. While I haven't used Google since they launched their AI features, I can say I have really enjoyed Kagi's AI responses to questions, and that each line of the response is referenced so I can check out the source (not sure if they all do this or not... I can only compare it to ChatGPT, which does not).
It's always a better use of your time to try something and see if you like it yourself, than asking strangers. Especially when there's no inconvenience or cost of trying. How would anybody here know what your preferences are in a search engine?
I'm not a Kagi user and don't have any hard info. But I find DDG frustrating with searches. It tends to do fine enough with technical searches, like software/programming searches that I make. It's other random searches that it often fails with. DDG will fail to find the nuance in my search query, the results will flat out not be what I'm looking for, so as a sibling comment mentioned I'll fall back to using !G with some regularity.
I've stopped using Google due to their captcha being so frequent and frustrating. I don't want to spend 60+ seconds answering 6 captchas purely because I use Firefox with privacy settings/extensions.
No Tor, no VPN. Nothing too far out of the ordinary.
I usually search in a private tab, enhanced tracking protection set to strict, uBlock with most default filters enabled.
Oh and a "Google Search Fixer" extension because the make the UI much worse for Firefox. Maybe it's one of my extensions, but DDG never traps me in captcha hell.
I would say that most alternate search engines don’t perform as well as google. Kagi performs better primarily because you can downrank or flat out block certain domains and also upvote helpful resources so that those domains are more likely to appear in your results. Also there are no ads so you get just the results and nothing else. When you sign up you even get a list of most blocked and downvoted sites from other users like quora is top blocked so it’s easy to improve your results. They also pay for access to certain databases and you get access to little things like court records when you search for people by name or other little improvements. I don’t think privacy is the primary reason to use it at all.
Honestly it’s awesome I use a software at work so I upvote that site and block the site of the competing software for example so I get the correct help guide. I don’t understand why google doesn’t offer many of these features.
It does its job: provides web results relevant to the query. At the moment it has acceptable signal-to-noise ratio (quality varies per-query, but it has higher chances of useful links than web spam), which is why people tend to say it's good.
After being a user for a long while, my enthusiasm for Kagi has decreased. Their UI is lovely, but I feel in the end they are just repackaging other indexers. I’ve started using Google + Ublacklist and for me it works the same.
I also don’t like how much they have focused on AI, given even their Quick Answer, when it’s wrong, it does so with such confidence it makes the tool quite untrustworthy.
I do think LLMs have their place in search and I think the Kagi approach feels a lot better than Googles'. Kagi doesn't inject LLM results anywhere, but they've been making LLMs accessible in their search interface for a long while - this being the most evolved version of that effort. I am not totally sold on everything they are doing but I hate their integration of LLMs the least.
Kagi's auto summary feature when you add a question mark after your query is absolutely excellent.
It essentially summarizes the top search results for you, leaning in on a strength of LLMs (summarizing) while reducing its greatest weakness (hallucinations).
For me, search integration of LLMs also helps bring together the value prop of paid search. It's not just privacy, it's the skillful integration of a premium information processing service (with built in privacy).
So for 25$ a month I get access to ChatGPT and Claude offerings in addition to access to Kagi search. This sounds like a good deal, compared to the 20$/month access to ChatGPT only. Or am I missing something?
You can use something like OpenRouter, which lets you access essentially all commercially available models. Including open-source models. There are no rate limits.
You pay a different rate per model (OpenRouter shows the pricing transparently). You load your account with credits. I use it daily (undoubtedly far more than the average user) and loaded 50$ with credits five months ago, but I still have over 1/2 of it left.
I think it is hard to believe that Kagi would be any cheaper and have no rate limits.
Keep in mind that, while OpenRouter gives you the upstream price for OpenAI/Anthropic models (so you pay the same per token), there's a loading charge, so if you want to load $10 in credits you pay $12 or so.
This means that it's more expensive than calling OpenAI directly, even though they have the same price per token.
Where the loading charge is amortized over all the calls made.
If you want to use precisely one API, paying directly for that API is cheaper. However, that's only true with closed-source providers. Anyone can host a server running llama 3.1 that OpenRouter could (in theory) use, bringing price competition to model cost. Closed-source models have a monopoly and can set their price wherever they want.
I'm okay with spending an extra 2$ every six months to access the APIs of any model I want.
LibreChat is also one of the few LLM chat interfaces that works with both external APIs (OpenAI, Anthropic, etc) and local ones (hosted via Ollama, etc) out of the box.
If it has a downside, I'd say its a little more involved to get setup, lots of docker containers, etc then a more batteries-included approach like Jan.
There's also BigAgi (really a weird ass name - probably hurting them) that is good for the same use case. Just paste in your API key and you get a really nice UI to chat through at-cost.
Also, if you're already a Kagi Pro subscriber it's really only $15/mo more for access to both models. This is the first time I've actually been tempted by one of these subscription LLMs.
It was a little janky when I tested it in beta and you don't get all the features of paying for ChatGPT directly (no multimodal, no DALL-E, etc) but otherwise yeah it's a good deal.
If you just want text chat with different models it's great.
What is the best paid service for private, anonymous, censorship free access to an LLM chatbot? Are there any that let you choose between multiple LLM backends to be able to compare answers or avoid being subject to secret system prompts, while still retaining privacy?
"Integration with Kagi’s legendary quality search results"
I don't disagree that this is useful, but I personally don't consider an assistant to be a chatbot that can tell me the weather. Assistants actively engage your daily life and do things that are usually considered tedious for people with a lack of time. Sure, that's a big ask for A.I in its current generation, but now for example I can ask Google Assistant (Gemini?) to save the shopping list I just gave it or even answer my calls in some cases. It's also certainly not the standard of human assistants, but it's closer than a chatbot.
I have been using Kagi for a while now. Something I have noticed recently is that it ignores a lot more of the words in my search queries, I felt the same thing with google over time, it shows results it thinks I want not the things I want.
Just using quotes usually was enough to make these words/phrases “required”. However, I believe there was a change (last year?) where they now sprinkle in other “relevant” results.
"Throwing the baby out with the bathwater" means trying too hard to do something good but accidently overdoing it!?
Just to make sure we're not accidentally giving Google a break, Google+ ended in 2019, if they cared for search at all they would have brought back this feature.
Yep, either it's gotten worse or they lied about not tracking your history, as it has gotten worse over time and I have found myself adding !g much more lately.
I don't know how to make this more meaningful than just an anecdote, but I love the idea of Kagi but just cancelled my subscription. All the issues about google search becoming more and more useless are absolutely true, but I still continue to get much better results for most topics with Google than Kagi. Same for Kagi's LLM products compared to directly using Claude or others.
I agree, but they aren't _that_ much better and I'd rather do a little more digging than support Google. I still find what I need and make good money. I can also rationalize it that I'm not always just taking the first result and actually reading multiple solutions to my problem but yanno, YMMV there.
My rule of thumb that I’ve found is that for the most part Kagi is on par with Google for anything non location-aware. Oftentimes it’s noticeably better actually since there’s less pumping of useless results to the top. For anything location-aware, however, it ranges from significantly worse to steaming hot pile of garbage depending on the search.
That’s what eventually made me cancel. Loved it for what it does give but found myself rather annoyed about having to switch to Google to get actual usable results on the significant amount of location based queries. Never got into the habit of using bangs unfortunately.
One feature here that I think competitors lack is that the LLM's view of search results can be constrained by Kagi's search "lens" [0] that let you exclude various categories of results.
I use Kagi but haven't dug into lenses, anyone have experience ?
I'm currently trying to write python script interfacing with outlook's mailterm interface (win32com.client) and it's annoying. I wonder if I can restrict search results to a particular domain so it only pulls from microsoft docs...
I use the Programming, Forums, PDFs, Recipes, and Small Webs lenses pretty regularly, though I haven't tried making my own lens yet. The 'Programming' lens is probably what you'd want for your Python script.
Note: I work part-time at Kagi (was a Kagi user before that), not doing search stuff.
> I wonder if I can restrict search results to a particular domain so it only pulls from microsoft docs
Yeah, you can set up an allowlist of domains in the lens configuration. I have a "Reddit" lens that essentially just adds "site:reddit.com" to the searches
I used the Kagi Ultimate subscription intensively for approximately seven months and am considering downgrading now.
While I am still happy with the search, I find the value of the assistant no longer worth it. At first, I thought I would be able to replace other providers with the wide range of offered models, but now I find myself often going to distinct providers. Kagi seems to have introduced a harsh and generalized input token limit for prompts (detached from what the underlying model can handle), which makes it almost impossible to work with code or documents. I think the VW example in the announcement makes it clear that the assistant is only intended to respond to prompts of a few sentences—searching by sentences instead of keywords. I don't see input for documents or things like code execution coming here, so the difference of $15 monthly can get me fancier features.
Mixing the output of the model with the limited content of approximately five websites (= Assistant with internet access), on average, is most of the time counterproductive for me. The sites mentioned in the sources of the output are often not sites I would usually visit or get information from (they usually don't correspond to the top results on Kagi web search). I have used search engines for decades now and have built a sort of index/pattern/feeling in my head for which site I am going to visit, which AI can not match for me.
The lenses are nice but are also pretty specialized. Most of the time I find myself just doing a broad query, which in most cases will already satisfy me.
I will definitely continue to use Kagi for search. There is no comparable search experience out there, and it makes searching definitely more effective.
I've been using Kagi for a while and I like it more that Google but I can't quite see myself paying for Kagi Assistant. I can't see myself paying for almost any pass-through AI subscriptions, I'd rather just pay for the core AI tool (ChatGPT/Claude).
This field is moving very quickly and since there will always be a lag in getting new features/models into the "wrappers" I'd rather get it from the source.
One random thing I wish Kagi could do is offer a way to promote the "official" website to the top of the search results. I have various up/down rankings applied but when I search for "DataDog" I want their official website at the top of the list, not under SO/Reddit/etc posts. If I'm searching for a problem then yes, I want those sites higher but it's slightly frustrating to have to scroll down 3-5 results to find the main website for a product/service/company. I feel like they should be able to differentiate between "search about X product" vs "X product".
You could just pin those domains as you come across them. I’ve definitely done that with one or two things that I hit this issue with.
Funny you mention Datadog, I specifically struggle to keep their blog/promotion/etc official results out of the way when digging around for documentation and troubleshooting. If I’m trying to find a solution to something weird with Datadog I’m often facing the opposite of your problem and wish Datadog’s site would get out of the way.
This is really cool. I'm very against AI in everything, so I probably won't use this, but I'm glad that they are making it an opt-in feature and that my $10/month plan doesn't go up to support it.
This is actually really exciting for Kagi. In a lot of situations the underlying model (Claude, GPT4 etc.) isn't that exciting, it's the connection to search to retrieve and summarize recent information that's exciting.
By already having a traditional search engine this puts Kagi at a big advantage compared to someone like Perplexity, or even Claude and OpenAI who I think are all cobbling together solutions on top of Bing's API.
The ability to use lenses with custom assistants is the killer feature IMO.
Want to search for open source projects that implement some algorithm? Create a Github/Gitlab lens. Want to ask it questions only about some framework? Add it’s domain to a lens.
I guess it depends on what we define as niche. I'm thinking for coding type problems as an example smaller less well known languages or frameworks that don't have extensive coverage across multiple websites etc tend not to show up well in AI Chat.
If you can focus the chatbot on a specific docs website with a lens though that problem might go away.
I've been using (and paying for) Kagi since the beginning.
For the first while, the search results were inarguably better than any alternative. I was thrilled and was recommending Kagi to anyone who'd listen. In the last six months or so, things have gone downhill. More often, my results ignore some of my search terms and I have to try to "trick" Kagi into matching them all. More often, I get only one or two pages of results when other engines give me notably more relevant hits. And more often, my search hangs or fails to return at all.
I don't care if they noodle with AI or make T-shirts as long as the search is great. What I've experienced is the search getting worse while they noodle with AI and make T-shirts, though. The two may be entirely unrelated but from an outside perspective, it feels like their core offering is suffering because of these other-than-search undertakings.
So far, I'm sticking with it, but my enthusiasm has definitely diminished. A couple of my friends have canceled their subscriptions in disappointment and I've started to consider cancelling mine, as well.
Is there some reason Brave Search isn't more popular on this site? It already has an incredibly helpful assistant built into the search and it's free. I don't even use Brave Browser (long live Firefox!) but I feel like people are sleeping on theur search.
Paid Kagi user here. I REALLY wish Kagi would focus on it's core selling point: search. Building a search engine is hard enough. I use Kagi Search everyday and I am mostly happy with it but the product has a lot of room for improvment.
Stop launching new products (browser, summarizer, gpt, assistant) while your core product is still behind the competition in many areas.
To me it seems like that's what they're doing here. I don't see right away how this is not their core business.
> Kagi Assistant has the ability to use Kagi Search to source the highest quality information meaning that its responses are grounded in the most up-to-date factual information while disregarding most “spam” and “made for advertising” sites with our unique ranking algorithm and user search personalizations on top.
Same. I love the AI additions and I think they've introduced them very thoughtfully.
To me, the best part of the AI additions is that it can (almost instantaneously) summarize information from the several top hits of a search. This is subtly but importantly different from having the LLM spit out an answer based on it's knowledge base, and also is able to quickly and easily cite it's sources! Extremely useful to me.
Agreed! To me LLMs as they stand now are a natural extension of the classic "Google". Which is to say a Natural Language -> Search Results list. People (and products lol) got hung up on LLMs returning the answers directly.. and while they've been a definite disappointment in that realm, they can be great for summarizing and aggregating imo.
I couldn't disagree any harder. When I can't find an answer, I turn to LLMs. I don't want to read half the docs on an AWS product, I want the snippet of code that I care about. Kagi, as best as I can tell, is the only search service which can answer these questions and also respects me as a customer.
I think of it this way: there's often not a single page (or even small handful of pages) that answer a query. The LLM features answer the question with text that links to the pages, rather than answering my question with pages that might contain pieces of answers.
The AI additions IMHO are search. Historically search gave us an ordered list of results, but there's no reason it needs to. The Kagi quick answer for example is phenomenal IMHO. Most of the time I am searching is because I need information for something. The "quick answer" and it's source citing can much more quickly tell me whether the results are worth a click. At this point I would hate to return to the old list of links output.
I think the problem is we're blending two different methodologies for finding information. When I search for something I want to get to the source. Others just want the answer. Ideally it should still be smart enough to figure that out, kinda what it does already if you search with a question.
If you hear Vlad talk it's very clear that he considers these things to be part of that core product. The summarizer powers the ability to summarize articles / search results (even video), the assistant and fastgpt power their answer to Google's snippets and quick answers, small-web is the minimum-useful thing to start their own index and not have to pay rent to Bing/Google, and they view Orion to be a long-term bet on the belief that this is the only way they'll get Kagi as a default search engine.
Great points. Though I think in order for any search to become a default in some browser app would take a billion or so in competing annual payment, or Google it will remain.
Also, AI aside, Searx has been around for many years as a very promising metasearch, even self-hosted engine, alas still little traction. Great to have all results, including re-ranked Google and Reddit in one place.
The move to AI stuff is why I don’t have a Kagi subscription. I really liked the idea of paying for search. I don’t want to give any money to an AI product.
I for one cancelled my Kagi subscription over this nonsense. I like paying for my search, but the AI "quick answer" integration is dangerously prone to incorrectly cited hallucinations, and I don't want to encourage this kind of irresponsible use of technology
I think this is very much about search. They just took on what Perplexity is doing with search. And I'm glad because I've been using it occasionally and now I can just keep everything in Kagi.
In my opinion, the results from Kagi Assistants are significantly inferior. I'm subscribed to both services and tried the Ultimate tier for a few days two weeks ago to test the new assistants. If they add the ability to search with multiple steps and allow the AI to digest more results, then they could be on par. It's more similar to ChatGPT at this stage, IMO.
The people I know who like to dictate into their phones and who have OpenAI's iOS app tend to open ChatGPT to "search" before they open Google or Ask Siri now.
They're going to go to one thing first, and this puts Kagi as an option.
Apple's alleged integration with OpenAI is presumably rolling in to Sherlock this though.
Why should they try to reach feature parity with some other engine that has more developers and more money? That's a losing strategy. They are running their own race and I think it's great. Kagi assistant is incredibly useful and there's no one else doing it like Kagi.
I think AI capabilities have been becoming an integral part of modern search. On the flip side you have the SEO optimizations.
Brave Search has offered an AI summarizer and assistant for a long time now. Bing with their OpenAI-powered Copilot. Google with the improved Bard/Gemini more recently. Amazon with the perhaps Anthropic-based Q for Business.
I think the end user is growing to expect the AI-augmented experience from all knowledge lookups. Feedback loop queries have become so natural to me, I’ve been finding it awkward to ask only one search query without a narrowing follow-up query, having the former discarded - kinda no longer adequate, particularly given the SEO-optimized flip side full of junk.
I agree that they should focus on their core product, but ironically while I use Kagi from time to time, I'm still mostly on DDG- via Orion, a browser I'm willing to pay for.
paying user here and i love these new additions and am super grateful they have added them. these tools are a real boon to improving search and for getting information.
No markdown blocks for code makes assistants really tough to use. It also seems to end up recommending that you search the exact same thing you ask it. For example, if I ask "name 5 websites similar to hackernews", it'll end up spitting out a link to a kagi search with "websites similar to hackernews".
I might end up disabling web search until they've tweaked it a little more. I really like kagi search so I feel pretty confident that they'll get it, but I don't think it's ready to replace the brand-name chatbots.
> No markdown blocks for code makes assistants really tough to use.
Could you clarify if you are referring to the markdown from your prompt not rendering? Or the response from the model not rendering markdown for code blocks?
The former was an intentional choice for now to simplify release; it is on our roadmap. The latter would be a bug - in which case, you can email the thread URL to zac@kagi.com and I'd be happy to take a closer look.
Isn't the point of a corporate blog to drive users to your product? Then why do the blogs never have an easy way of getting there? Clicking the logo and things in the header all just take me to the front page of the blog. Pet peeve of mine.
I think it is because bearblog (software we use, or at least the version we have) does not allow custom external in the navigation. But the product website is linked to in the very first sentence, and also from the home page of the blog. Still not good? :)
I honestly think FastGPT is the best implementation of AI w/search, and is extremely versatile/useful across domains (granted I don't code). I think it's the same thing as the Quick Answer feature from their standard search.
In my daily work as an MD it's become my reflexive go-to for looking up answers to specific to general, easy to complex clinical questions. I use it far more frequently than UpToDate (which is no less than the holy book of medicine), more than PubMed/Google Scholar searches, and definitely more than a basic web search (Google, only b/c it's a hassle to log in to Kagi every session at work).
Maybe 1 time out of 10 it won't give a correct or meaningful answer (in which case my prompt needs to be refined, or is just not suited for this kind of tool). But apart from that it will give me exactly what I need, because it uses Kagi search to source its answers. Kagi search does a decent job bringing to the top relevant journal articles (which in turn may mention other articles, adding indirectly to the trove of sources FastGPT pulls its final answer from). It shows the 5 search results it referenced at the bottom of the page, so more often than not if I don't get my answer in the direct summary, I have very relevant sources to read through.
I also don't think you need a Kagi account to use it.
Something I love about Kagi that isn't often known, is they will pro-rate based on days. If you are on an existing plan, you can upgrade to Ultimate, try it out for a few days, and then downgrade and only pay for the days you used it. I despise the subscription model generally, but if we're going to have it then I wish more companies would do pro-rating! Anyway, you can try it out for very low risk.
I've been curious about Kagi as a search engine for a while now and this seems like a good time to try, given that I already pay $20/month for ChatGPT.
The thing stopping me currently from trying this or Claude is I rely on the Opt+Space shortcut with the ChatGPT mac app.
Are there any other options for a native mac app with integration as good as the ChatGPT app?
I use Raycast which is a drop-in replacement for Mac spotlight along with the ChatGPT extension that lets you quickly search conversations / ask questions / etc once you provide it with your OpenAPI key.
I have been paying for Kagi for the past five or six months and apart from the truly niche cases where I need top of the shelf search results the majority of the time the top results are not that much different from, say, duck duck go. My other issue is that I am on the first subscription tier which currently supports 300 searches and I often find myself reaching that limit sooner that I would like. I will continue to pay for the service, but so far the, admittedly, better results haven't given me this light and day experience other users describe. Maybe I am just not using the engine to its full potential...
agreed. I'm getting pretty sick of AI summaries that will fully lie to me about things. Google's search summary AI is awful I'd expect Kagi's to be similar.
I think you should try Kagi before passing judgement, really clever design.
You only get the quick summary if you ask for it (i.e. you're happy to take the risk). Another handy use, is web summarize, which looks at the specific webpage, lowering the risk even further.
If you don't call these features it's straight up clean search with no ads and it's fast with good results.
> You can edit the question and add that you’re working on a binary classification problem to get a more specific answer.
This mid-thread editing feature sounds really useful, I'm curious how does it work when you switch between models in the middle of a conversation?
Like, say I start with a general search question, then halfway through I want to switch to a coding model to ask something like, "Can you create a Python dictionary of the top 10 longest city names in the UK and their populations?"
Does the context carry over smoothly, or would I need to rephrase things when switching models? Wondering how it handles tasks that require different kinds of expertise without losing track of the flow.
I've been trying out the new Android app. I know it's still beta, but it sucks big time. I'm a fan and paying user of kagi, just wanted to point that out in case any kagi developer is reading :)
among other things, it randomly says I have no connectivity in an awful modal, I click to search and the keyboard doesn't open, I hit "enter" to search and nothing happens (I need to tap on the search suggestion). pretty bad so far. will try it again in a couple of weeks to see if there is any progress...
Sourcing information from better search results. I cancelled my Perplexity Pro since for any use case I had for it, I would instead use Kagi FastGPT. I tried Assistants (beta) but I didn't think it was anything special, didn't really see a way to integrate it as a daily tool, and ultimately FastGPT gave me the best answer, even better than their gpt4o and Claude Opus/Haiku based assistants.
I love Kagi and happily pay their $25/mo but I think it's a mistake to think of their offerings as cutting-edge AI. It's obviously limited compared to open source software (as mentioned elsewhere in this thread already) and likely more expensive than raw API calls. This isn't the "best" AI experience.
What it is though, is fast, available on all my devices, constantly upgraded, and integrated with their already excellent search engine.
When I see these sorts of announcements and read some of the comments here, it makes me worry that bad customers cause enshittification and I hope kagi stays true to their human-friendly web search product.
It's not clear from the post how to access this new assistant. The search page has no such option (not a paying user). When I run a search, I only see an LLM based summary of the results similar to Google's.
Update: I see now that they say it's not available for free users. Need to pay $25/month. Not sure why, they can offer it for free users with the cheaper models like they do now to generate a "quick answer". I'm not going to pay to try it out.
I recently tried Kagi and I struggled to see the value.
For many queries side-by-side with Startpage it delivered the same results word-for-word (sure you get a few sponsored links top-3 of Startpage but its no big deal to scroll past those).
For other things, it was just plain annoying, e.g. "newest $type restaurants in $large_city" half the results on the first page were from 10 years ago (e.g. dated 2014). I mean FFS I put the word "newest" in there !
They seem to have a habit of interespersing very weird Facebook links randomly in the middle of a list of results. For example I was searching for something related to a specific Prometheus function (which I explicitly named in the query, alongside the word prometheus) and Kagi insisted on interspersing the technical results with random links to Facebook pages of companies selling "girlie dresses for proms".
I approached Kagi with an open mind, but having used up the 100 free searches nothing made me say "just shut up and take my money".
For some of us it is. If your search engine's revenue model is based on advertising to its users, their relationship is fundamentally adversarial. This affects all of their decisions, in ways that are sometimes hard to identify. Witness the slow decline of google search result ads.
If users are the direct source of revenue, then everyone's interests are aligned.
Also, I, and many like me, value a lack of ads much more highly than you do. Which is fine.
I don't think it's necessarily the case that interests are aligned just because users are the direct source of revenue. Many services that rely on users as a direct source of revenue still have interest alignment issues for various reasons.
For example, I don't want to log into or provide payment information to my search provider, because I don't want to voluntarily provide personal information. Compared to Kagi, I can use traditional ad-based search in a relatively anonymous way.
Avoiding the ads doesn't fix the alignment issues. Even without ads, modern google search is dramatically worse than a decade ago, and I'm personally pretty confident it's because their interests and user interests are not well aligned. I don't have to worry about that with Kagi.
Customizing your search results is a big part of it. When I tried Kagi, I did not find it to be a huge improvement on Google until I started adjusting the rankings of my search results. Now I find it painful to go back to Google when I use someone else's computer or device.
The other big part of it (for me at least) is seeing more obscure websites in my results. I have had Kagi for a year now, and it has saved me more money than I've spent on it by making it easier to find specific products at lesser-known shops. These lesser-known shops often have really great sales because they are trying to compete against the big names, and Google pretty much only shows me the big names.
Hey folks, how does Kati compare to perplexity.ai ? I have been using the latter and seems like I can’t go back to anything else. I did use magi for a while but perplexity seems crazy perfect for search and quick summaries a
LLMs + search are really useful. I use phind.com regularly for this, it is remarkably good for enhanced search queries. I use Claude a lot for more general knowledge stuff but its inability to provide references or do web search like stuff holds it back.
What’s the incentive for websites to let Kagi and others indexing content if llms in search show relevant informations right away? Wouldn’t something like perplexity ai making more sense then? Or perhaps better application of llms to search
I haven't used Kagi's Quick Answer very often yet, but when I do, it always cites its sources and I often end up clicking into at least one of the sources to look for more detail or context.
Bingo. I nearly always use the quick answer, and then will use the cited sources to click onto the page to either read more or verify that the summary was accurate (and it always has been in the 300+ times I've used it).
If you publish to make information known, then your incentive is that it helps spread that information to people who may not otherwise visit your site. If you are trying to make money off of search engine traffic then you might not like this much at all. I think most people would rather not be pointed to those sites in the first place, so it’s a win-win if they block crawlers.
This is a bad take. I don't make money off search engine traffic (beyond the occasional donated dollar), and yet I would quite rather[0] that AI doesn't visit my site.
Imagine for a second a world where instead of publishing directly to your own website (home), you publish into the LLM's knowledge base. Nobody comes to you for information, they come to the LLM. Nobody cares about you, or the work you put in. Your labour is just a means to their end. To some extent, you could argue that Wikipedia works in this way, but it really doesn't. The work you put in is reproduced verbatim, and the work is collaborative. You get the joy of seeing your writing being used to help other people in a very direct sort of way, as opposed to being aggregated, misinterpreted and generally warped by a non-intelligent LLM.
In other words, you cannot possibly expect others to want to work in a sweat shop, toiling away to provide you with instant gratification. We must leave room for human expression.
Okay, I’ve imagined the world you described. I don’t care for it. I also don’t think that’s a likely outcome of LLMs. Why would somebody continue “toiling away to provide you with instant gratification” if there’s nothing in it for them?
I've been using Kagi for a while (almost two years now!) and it's been nothing but excellent!
Lenses are very useful (Reddit lens is on every second search), and I personally really like the AI features they are working on.
The new more advanced assistant which is able to do searches, which can also be constrained to lenses, and lets you pick an arbitrary model, is excellent, and basically means I don't need a chatgpt/claude subscription, as Kagi covers it very well.
All in all, great product which I'm happy to pay for.