Kagi vs Perplexity: price, privacy, and models, with Morse as a third column

The Morse teamPublished November 2, 2026Facts checked November 2, 20268 min read

Kagi vs Perplexity comes down to who pays and what is kept. As of November 2, 2026, Kagi is a paid, ad-free search engine whose AI Assistant never trains on your threads and deletes them a day after the last message by default, from $5 a month. Perplexity is an answer engine with a free plan that keeps your history while your account is open and, unless you opt out, uses your searches to train its models, with Pro at $17 a month billed annually. Choose Kagi if you want search you pay for and less kept about you than Perplexity keeps, Perplexity if you want the most research tools per dollar, and Morse if you want AI chat with web search and no chat history on Morse's servers.

We read Kagi's pricing page, privacy policy (changelog entry September 22, 2026), FAQ, Assistant and Privacy Pass docs, search sources page, and changelog, and Perplexity's Pro and Max pages, privacy notice (last updated October 23, 2026), and seven of its help center articles, all on November 2, 2026. Every Kagi and Perplexity fact links to their own page. To their credit, both say in their help centers that the model providers behind their answers can't train on what you send.

Kagi vs Perplexity at a glance

As of November 2, 2026.

Is Kagi or Perplexity better for privacy?

Kagi, on every count its own pages cover, starting with who pays. Kagi's FAQ says "We're entirely user-funded," and that it needs an account "only because it is a paid service" (Kagi). Its privacy policy says search queries "are only logged temporarily for debugging purposes and are automatically purged after a short period," and that Kagi doesn't "load any analytics or telemetry" (Kagi).

Perplexity lists "AI model training" among its reasons for collecting your data, "unless you opt out through Account Settings" (Perplexity).

Each service's settings in full are in is Kagi Assistant private and is Perplexity private.

What happens to one question in Kagi and in Perplexity?

Say you ask both the same question, signed in on a paid plan with the default settings. This is where it goes, by each company's pages read on November 2, 2026.

Perplexity's help center gives two answers about Incognito. Its "Where did my sessions go?" article, last modified October 27, 2026, says incognito sessions "expire after 24 hours," and further down that in Incognito "your history is temporarily hidden from the sidebar, however all your sessions will still be available to you under the History page" (Perplexity). The same article says "Perplexity doesn't delete your data" to explain missing sessions. If you rely on Incognito, check your History page after a session to see which answer holds for your account.

Which is cheaper, Kagi or Perplexity?

Kagi costs less to start, and Perplexity is the one you can use free. Perplexity's own Pro page showed $17 a month billed annually on November 2, 2026, and Max at $167 a month billed annually, each with bonus credits for its Computer agent (Perplexity).

Kagi's annual plans save 10% (Kagi), and its AI use is capped by value: "a $25 monthly plan covers up to $25 of token usage," with a 20% markup on token costs (Kagi). Perplexity's model access "may be limited during weeks of especially heavy usage" (Perplexity).

Which has better AI models, Kagi or Perplexity?

They share most of their flagship models. Perplexity's model table, dated October 27, 2026, gives Pro GPT-6.1 Sol, Gemini 3.8 Flash, Claude Sonnet 5.5, Kimi K3, GLM 5.3, Grok 4.7, and Nemotron 3.5 Super VL, and adds GPT-6 Astra, Gemini 4 Argon, Claude Opus 5.5, and Claude Fable 5.1 on Max (Perplexity). Free users can't pick a model. Kagi Ultimate lists all of those flagship models except Nemotron 3.5 Super VL, plus DeepSeek V4 Pro, Qwen3.8 Max, Mistral Large 4, and others (Kagi). Kagi includes Opus and Fable at $25 a month, while Perplexity keeps them for Max.

Perplexity builds more around the models. Max adds Model Council, which consults several models at once (Perplexity), and Brain, a memory system for its Computer agent, in research preview (Perplexity). Kagi's Assistant takes uploads up to 30 MB and can share a thread by link (Kagi).

Does Kagi or Perplexity show ads?

Neither shows ads in its answers. Kagi says it has "no ads, no tracking, no noise" (Kagi). Perplexity's cookie notice says it "does not use these technologies to sell third party ads on our services" and uses them "to measure our ad performance on third party partner websites" (Perplexity). Its privacy notice's California table still has a column for "Sharing for Targeted Advertising" (Perplexity), which you can opt out of through its cookie pop-up.

Where Kagi or Perplexity is better

• Kagi is better for search itself. Lenses and personalized results carry into the Assistant's searches (Kagi), Safe Search now blocks known scam sites (Kagi), and Privacy Pass can unlink searches from your paid account.
• Perplexity is better for long research jobs. Deep Research "performs dozens of searches automatically" and returns a report in two to four minutes (Perplexity), and Model Council, its Computer agent, and the Comet browser have no Kagi equivalent. What Comet collects is in is Perplexity Comet safe.
• Unlike Kagi, Perplexity has a free plan that doesn't run out. Kagi's trial ends at 100 searches.

How Morse keeps your questions off a server

In both Kagi and Perplexity, an AI thread sits in a database under your account. Kagi deletes it sooner and never trains on it. Perplexity keeps it longer and trains on it unless you say no. Morse is built so that copy doesn't exist.

Your chat history is stored in your browser, encrypted, and Morse's servers don't keep messages or log what you write. The account holds your sign-in details, plan, balance, and a usage ledger with no content. Morse doesn't train AI on your content, sell your data, or show ads, so there's no "AI data retention" switch to hunt for. You get over 30 models, open and closed, in one account, and web search comes with the Free plan. Then pick a privacy mode for each chat:

• Anonymous, to ask GPT, Claude, or Gemini without the provider knowing it's you. Our proxy removes your account, email, and IP address and sends the question under Morse's own account. The provider may keep it for a limited time to check for abuse, under its own policy, so keep names out of anything sensitive.
• Private, for open-weight models and Grok with nothing kept. Private is the default for a new chat. Morse has zero data retention agreements with its inference, search, and other providers, and where a provider confirms zero retention on each reply, Morse checks it and refuses a reply that doesn't.
• E2EE, for the questions no one should read. Your browser checks the enclave's hardware attestation, then encrypts the message so only that enclave can read it. Morse's servers relay ciphertext they can't read. E2EE runs open-weight models like GLM, Kimi, and DeepSeek, has no web search, and rests on the enclave hardware and attestation working as designed. End-to-end encrypted AI chat shows the check.

Because your chat history is stored only in your browser, encrypted, clearing the browser or deleting your account removes it for good, and Morse can't recover it. Copy out anything you want to keep. And because we don't store chats, we usually have very little to give authorities.

Choose Kagi, Perplexity, or Morse

• Choose Kagi if you want a search engine you pay for, with AI on top that never trains on your threads, and you'll use Privacy Pass to keep searches apart from your account.
• Choose Perplexity if you want a free plan, or Deep Research, Model Council, and an agent, and you'll turn off "AI data retention" on day one.
• Choose Morse if you want GPT, Claude, Gemini, and open-weight models in one account, with web search, history that stays encrypted in your browser, and end-to-end encrypted chats for anything no one should read.

Ask your next question on the Morse home page without an account, or start free with no card. Pro is $19 a month, or $17 billed yearly, and adds E2EE chats. More options are in Perplexity alternatives.

Morse publishes this comparison. We've tried to keep it fair; if something's wrong or out of date, email hello@trymorse.ai.

Kagi is a trademark of Kagi Inc. Morse is not affiliated with or endorsed by Kagi Inc.

Perplexity is a trademark of Perplexity AI, Inc. Morse is not affiliated with or endorsed by Perplexity AI, Inc.