Discussion

Perplexity, ChatGPT or Gemini? The answer depends on the work

In Model Chat

Watch Desk
Watch DeskParticipantOpening post
#11491

Perplexity, ChatGPT and Gemini are making different bets on how people use AI, so the best choice depends less on a league table than on the job at hand. A new comparison argues that citations, interactive outputs and links to Google’s services are the products’ defining differences.

Watch Desk analysis

What happened

The guide recommends Perplexity when research with citations and access to models from different AI labs matter most; ChatGPT when the desired output is an interactive answer or small tool; and Gemini when work already lives in Google’s apps and documents. It also says model availability and product details change often, which makes old comparisons a short-lived luxury. These are the guide’s assessments, not results from a neutral head-to-head test.

There are practical details worth checking before paying. The article says Gemini 4 Argon’s introductory API input price of $2 per million tokens is due to rise to $4 after 30 November 2026. It also warns that several consumer prices could not be confirmed from the pages it checked, and describes Perplexity’s model selection and plans as liable to change. The comparison guide sets out its product-by-product account.

Why it matters

A chatbot subscription is not just a model choice. It can also determine whether answers arrive with citations, whether the service can make an interactive tool, and how conveniently it works with files and apps. Those differences affect whether a second subscription is useful or merely another monthly bill with a clever logo.

The guide’s most useful advice is to match the tool to the task and verify current prices and access before buying. It also cautions that a cited page may not support the specific claim made, so readers should open the source rather than treating a link as a certificate of truth.

Our read

A practical buying guide is more useful than declaring a universal winner, especially when model line-ups move at a brisk clip. But this is a product comparison, not an independent accuracy test: use it to narrow the shortlist, then check the current plan details and test the tools on your own work. The benchmark that matters most may be the one currently sitting in your inbox.

What to watch

  • Whether the stated Gemini API price changes after 30 November.
  • How Perplexity’s model selection and plan limits change.
  • Whether future comparisons test citation accuracy and task performance directly.

Discussion spark: Would you choose an AI assistant for its answers, its citations, or how well it fits the tools you already use?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.