Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Kagi is contemplating the removal of the assistant from its professional tier

kagifeedback.org

Post reply on HN
Print view View original

71–80 of 110 posts

Re: Kagi is contemplating the removal of the assistant from its professional tier

#71

Post by willis936 »

mpalmer wrote: post #31
leokennis wrote: post #17

I admire and applaud Kagi for what they are trying to be. But on the AI front, the Assistant is simply worse than using for example Gemini or ChatGPT directly. It is slower, it cannot generate images etc.

On the other hand, it doesn't monetize any of the data you give. Six of one, half dozen of the other, yeah? Re: image gen... it's a search engine. Why would I need my refrigerator to toast my bread?

The privacy buffer between the LLM providers and the user is a large part of the appeal. Having it hooked up to a search engine outside of the data broker space makes it uniquely attractive.

Given that window dressing, having toasted bread in a kitchen that isn't selling my data is something I want.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#72

Post by genthree »

EbNar wrote: post #9
sph wrote: post #5

Good. I'm a paying user from day one, and I never used, nor want to see any AI features on my search engine. I'd use Google if I cared about any of that stuff. Honestly, I just want my money to be used to improve good old match-keywords-against-index search.

Well, good for you, I guess. But for the rest of us who use the Assistant this would be a net downgrade, AKA paying the same for less.

Is it any better than Google's or the one in DDG? Those things are wrong most of the time for questions that couldn't have been answered by looking at a few words in the summary of the very first search result anyway. I find they only save me time if what I was searching was so trivial that it doesn't matter at all if it's wrong ("How old is such-and-such actor?") and I didn't really even need to search it in the first place. Otherwise, if it actually matters, I still have to check the sources, because no matter how confident the answer looks, it's wrong way too often to rely on it.

(Gather 'round, kids: used to be, questions of utter trivia like the exact ages of actors or whatever didn't even take root in our minds in the first place, we didn't even feel the itch to answer them to begin with, because it was far more effort to answer them than it was worth, unless it was a topic you cared enough about to have a bunch of books about it on-hand, so such questions would usually fall out of our brains before we even consciously engaged with them. Our heads were a lot less noisy then. It was very nice.)

Re: Kagi is contemplating the removal of the assistant from its professional tier

#73

Post by omnimus »

senko wrote: post #42

I use Kagi search all the time. I often use the Quick Answer feature (append "?" at the end of the query, gives an AI overview/answer, then standard search results below). I never use the full Assistant - I have ChatGPT and Claude for chatbot use cases. I like Quick Answer because of its ligtweight UX. Ctrl-T for a new tab in any browser window, type the question, get the answer. This is faster and mentally lighter t…

You can add some "bang search" (like your ? for quick answer) to use the other models Kagi supports like Kimi.

The quick model is there to be mainly quick.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#74

Post by omnimus »

treesknees wrote: post #27

I’ve used it quite a bit, but just as a natural language frontend to their search engine. I don’t see it as a true alternative to generic chat bots like Claude or ChatGPT, and certainly not a coding tool. I would likely not subscribe to Assistant.

You can look at it other way around - why to have ChatGPT when you can use chatgpt through kagi. One of the advantages being that they claim to be "privacy" layer that doesn't give other companies metadata about you.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#75

Post by EbNar »

genthree wrote: post #72
EbNar wrote: post #9

Earlier quoted context omitted.

Well, good for you, I guess. But for the rest of us who use the Assistant this would be a net downgrade, AKA paying the same for less.

Is it any better than Google's or the one in DDG? Those things are wrong most of the time for questions that couldn't have been answered by looking at a few words in the summary of the very first search result anyway. I find they only save me time if what I was searching was so trivial that it doesn't matter at all if it's wrong ("How old is such-and-such actor?") and I didn't really even need to search it in the fir…

I've never user Google or DDG assistants, tbh. My use for Kagi's assistant is for stuff related to work. I'm a university prof., so it "saves me a bit of time" while checking for errors in assignments, clarify concepts to explain to students and so on. I almost never use it to search for stuff.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#77

Post by nchagnet »

DamnInteresting wrote: post #68
zetanor wrote: post #62

Strangely: "Searches are, unexpectedly, more expensive for us to serve than AI. A user that does only traditional searches (not using Assistant) will cost more on our end than someone using AI + search." https://kagifeedback.org/d/1338-provide-a-plan-without-ai-fu...

> A user that does only traditional searches (not using Assistant) will cost more on our end than someone using AI + search. Unless I am missing something, that appears to be a mathematically impossible claim. They are saying A > A+B, where both A and B are positive values. I suppose it could be that non-AI users do more total searches, and the extras add up. Hmm.

They're likely saying that at equal usage, the user with mixed usage will cost less because the cost of B is lower than A.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#78

Post by buster »

leokennis wrote: post #17

I admire and applaud Kagi for what they are trying to be. But on the AI front, the Assistant is simply worse than using for example Gemini or ChatGPT directly. It is slower, it cannot generate images etc.

Keep in mind that kagi offers a wide range of models, not just one. I wouldn't want to have multiple subscriptions (for chatgpt, anthropic, gemini etc.)

Re: Kagi is contemplating the removal of the assistant from its professional tier

#79

Post by i_love_retros »

leokennis wrote: post #17

I admire and applaud Kagi for what they are trying to be. But on the AI front, the Assistant is simply worse than using for example Gemini or ChatGPT directly. It is slower, it cannot generate images etc.

I see the slowness and now the news that they might remove the assistant for the $10 plans as evidence of how costly it is to run LLMs and by extension how unsustainable it must be for OpenAi, anthropic, Microsoft etc to be offering such performance for free or very low prices. Surely something has to give soon.

Re: Kagi is contemplating the removal of the assistant from its professional tier

#80

Post by 9Ljdg6p8ZSzejt »

oidar wrote: post #58

Ultimate user here. I assume this doesn't kill "quick answers" for pro users - which I use frequently when I need a quick summarization. For assistant use directly, I've been thinking about stepping back from ultimate, as I use claude for AI rubber ducking; which works better than all of the LLMs available on assistant.

I have maxed out my ultimate usage before and when that happens the quick summarization tools did not function indicating I hit my limit. I assume it would affect those, but that might be part of how they break it up.
Post reply on HN