Hacker News

points by mzhaase 2 days ago | 0 comments

The thing I use most is kagi AI assistant. They have a bunch of models like Deepseek 4.1 and I don't think they have any usage limit.

[–] _nalply 2 days ago

They have. Look in your account, it's in usage. If your usage exceeds what you have paid you will get told.

This said, it's a good offer. You can switch between many models quickly. They are rated by quality, speed and price, and it's shown with orange dollar symbols for price and bullets for speed and quality. Example:

    Anthropic Claude Opus 5.5

    Anthropic's flagship model with always-on adaptive
    thinking ($$$$$). Best for deep reasoning, agentic
    coding, and long-horizon work.

    Cost    $$$$$
    Quality ●●●●●
    Speed   ●●

[–] darkamaul 2 days ago

I have quite the opposite experience here where I almost never use the assistant. I much prefer the frontier labs alternative (I’m lucky enough to have my work pay for the subs).

[–] ctolsen 2 days ago

They do, it’s capped at whatever your subscription cost is. But if you don’t have Ultimate the models are cheap and it’s hard to spend it all on assistant use.

[–] SparkyMcUnicorn 2 days ago

I've come close to my limit a few times, and I'm sure you could hit it if you're using expensive models a lot, but it's still pretty generous.

For those that are looking for it, you can track your current month's billing/ai-usage here: https://kagi.com/settings/billing

[–] insane_dreamer 2 days ago

if I'm using Kagi it's because I don't want AI replies. If I wanted AI, I'd use Google/Gemini (which is arguably better at this particular type of task than the open models).

[–] p1necone 2 days ago

I've been kinda disappointed with the $10 plan tier of the AI assistant. It looks good at first glance - shows you that it's doing a bunch of research, you can see which sources it's looking at etc. But then when it goes to report results on anything data heavy it makes a lot of mistakes with putting figures under the wrong headers, using incorrect values etc.

The last time I used it heavily was a few months ago so it might have gotten better, and I assume the more expensive tier where you can choose which models it uses might be a bit better depending on which model you choose. I much prefer just pointing my own local agent harness at a search api at this point though.

[–] Matumio 2 days ago

If you mean the "Quick" assistant, I only use it for light queries but I think it's pretty accurate usually, especially if the search results have the answer somewhere. It sometimes takes your query too literally, like when using quotes it will search with quotes, find nothing, and search again with quotes. Which is silly because if I wanted literal search, I wouldn't turn on the assistant!

But since I'm on the Ultimate plan, for stuff that matters I usually use Fable with a custom assistant (prompt), and sometimes the stock "Research" assistant. Both work really great for me. I use the "Research" assistant for more technical questions, it tends to dig up information from forums and cross-reference it with the source-code file it digs up from github and quote the RFC that it has directly fetched too.

Though I usually prefer my custom assistant prompt, which Kagi seems to combine with their general prompt in the background. This works really, freaking great most of the time. My custom prompt is an elaborate version of "answer the question, but then try to second-guess what the user is doing and also provide a different perspective that could be more useful than a direct answer". This has derailed my narrow train of thoughts a few times and turned me around into the right direction.

That said, I haven't played around with my own local agents for search yet to compare (in part because Kagi does a very good job at this), and this is something I want to try, in a "own your tools" sense. But the first thing I'll make sure that it uses the Kagi API for its search.