Kagi Assistant

“It is very capable. I use the Kagi mcp with coding agents and I like the results”

“Their assistant tool is already quite good - it has replaced search for me in a number of scenarios.”

“It doesn't just give a summary, it is more like a personal "Googler"/web surfer. I feel like I'm involved in the process.”

“Kagi Assistant defaults to the last-used model unless you got there from a Quick Answer”

“you are probably better off using Kagi Assistant anyways, as you get $10 there on the $10 Professional plan”

“I've run into two instances in the past week where it confidently gave me wrong information”

“tends to produce higher-quality results for information that would not normally be within a given model's training data”

“Yes, I use AI assistant a lot. I copy the agent definition from Geir Isene to a custom assistant, it works nicely.”

“it's incredibly useful. It misses the mark frequently, like any LLM, but it's well annotated with footnotes that link to the original material.”

“it often knows when there are alternative names for the same sort of product and so can find things that are rebranded”

“Personally I think this will replace other LLMs for me, at least for QA usage.”

“their LLM assistants are actually better for my purposes than Anthropic or whatever”

“That plan comes with Kagi Assistant, which is more than enough with big selection of simple models for a great portion of AI assisted queries.”

“gives you control over what context from the internet the AI can use/you trust”

“I use these a lot. If I find I keep front-loading a lot of context I just create an assistant.”

“It's insanely fast and seems to follow instructions quite well as a research assistant.”

“Kagi feels more like the results I would have had from ChatGPT a couple years ago”

“Kagi has it as an option in its Assistant thing, where there is naturally a lot of searching and summarizing results.”

“For these uses, LLMs are wonderful — and Kagi Assistant is plenty good enough.”

“has worked great for me in scenarios where the search result data is more important than the model”

“Kagi Assistant works pretty well. Its version of Claude Opus is much smarter than the one that I get with a plain Claude Pro subscription.”

“the deep search integration is precisely what makes this specific AI assistant better than, for example, ChatGPT”

“One of the advantages being that they claim to be "privacy" layer that doesn't give other companies metadata about you.”

“Adding a LLM on top of search results and to fire off multiple searches makes it possible to sometimes squeeze a little bit of actually useful information”

“It uses neutral system prompt by default, whereas Gemini and ChatGPT will please you too much”

“the Assistant is simply worse than using for example Gemini or ChatGPT directly”

“It's more of a glorified search summarizer but it's essentially free with my existing subscription.”

“Kagi Assistant helps a lot in this regard because searches are ranked using personalised domain ranking.”

“The key is causing the model to increase your friction (causing learning and memory) instead of decreasing it.”

“thanks to the strict privacy conditions of the API and the mixing of queries from thousands of users”

“I still felt that there was still very much a privacy trade off”

“its really good (GLM 4.7) but still there is some amount of tracking/logs still kept”

“the most useful tool I've found as far as searching goes, and occasionally simple codegen”

“Time spent isn't always the best tell of quality, particularly when the search engine used returns little noise”

“I'm loving the features that are coming out. I use Translate, Assistant, Universal Summarizer, and (of course) Search multiple times a day.”

“It's so convenient, I've implemented identical feature for our product at work.”

“They support many providers and new notable models are typically rolled out only a few days after release”

“Kagi actually got those two correct, explaining how losing your king means losing your chess game”

“I don't see much difference in information compared to using existing models”

“Even today more that 80% of our members use Kagi Assistant and our other AI-supported products”