Introducing Kagi Assistants

Re:Introducing Kagi Assistants by nsonha on Thursday November 20, 2025 @09:34PM (#46000606)

I use perplexity a lot and pretty much exclusively with "deep research" on. Is this on the same level? Because Perplexity often take more than a minute, and this is only 20 secs.

[ Parent | on HN ]

Re:Introducing Kagi Assistants by milch on Friday November 21, 2025 @01:37AM (#46001825)

A query I ran earlier ran for about 70s, and did 12 web searches and 2 site fetches.

[ Parent | on HN ]

Re:Introducing Kagi Assistants by Havoc on Friday November 21, 2025 @02:06PM (#46007733)

An employee above said they’re using cerebas so time may not be a good measure of thinking effort here

[ Parent | on HN ]

Re:Introducing Kagi Assistants by Rehanzo on Friday November 21, 2025 @07:27PM (#46010782)

For most queries it is around the same level. Time spent isn't always the best tell of quality, particularly when the search engine used returns little noise

[ Parent | on HN ]

Re:Introducing Kagi Assistants by nsonha on Saturday November 22, 2025 @05:13AM (#46013609)

I don't think the reasoning level is what makes it for me. I am treating this as a search engine. I want the thing to fetch/read from index as many pages as possible. Either that or as many agentic tool calls as needed. That takes time.

[ Parent | on HN ]