Duolingo vs Rosetta Stone: the prices, checked today

Rosetta Stone costs $19.99 a month, or $159 billed yearly. I read that off rosettastone.com/buy this morning, 21 September 2026.

The page ranking first for this question today says Rosetta Stone is "$10 to $15 per month." Another quotes $143.88 a year and a $299 lifetime plan, then admits in its own parentheses that the figures are from November 2022 while carrying a 2024 title. The third never states a Rosetta Stone price at all.

So a reader with a credit card open and a real question gets three answers, none of which match the buy page.

Who I am and what I'm selling

I build Verbamor, an iOS app that turns vocabulary from your own tutor lessons into flashcards. It costs money. I have an obvious reason to want you unhappy with both products on this page.

I also haven't used either one. Everything here comes from Rosetta Stone's and Duolingo's own live pages, quoted with the date I fetched them, rather than from a personal trial I'd have to ask you to take on faith.

That's a weaker claim than "I tested all 14 apps for 12 years," which is what the competing pages say. It has the advantage of being checkable. Click any source at the bottom and you'll land on the company's own words.

Two of the three pages ranking above this one run affiliate buttons next to their verdict. One of them recommends Rosetta Stone in a scenario where it has just finished explaining that Duolingo is better for that reader. I'll be recommending against myself twice before the end of this page, and once I'll be sending you to the free option.

What they actually cost, checked today

Rosetta Stone's buy page, 21 September 2026:

• $19.99 a month, monthly billing, all 25 languages.
• $159 billed yearly, which the page displays as $13.25 a month with a "SAVE 33%" badge, all 25 languages.
• Both carry a 14-day money-back guarantee. Rosetta Stone's support site puts it as "All Sapphire subscriptions purchased directly from Rosetta Stone come with a 14-day satisfaction guarantee."
• There is no lifetime plan on that page. If you've read that Rosetta Stone sells lifetime access for $299, that's a product the buy page doesn't offer today.

Two things sit outside those numbers, and both get missed.

The first is the free trial. Rosetta Stone's support page describes "this one-time only 7-day free trial," and it's blunt about the end of it: "Your Rosetta Stone Sapphire account will automatically convert to a paid monthly subscription once the free trial ends." Monthly, at $19.99, so the trial's default landing spot is the more expensive tier. Set a reminder for day 6.

The second is tutoring, which the membership page lists alongside the lessons as though it were included. It isn't. Rosetta Stone's own support page prices it separately: "5 sessions for $279" or "10 sessions for $399," each one "30 minutes long and delivered via video call on the Wyzant platform." That's about $40 to $56 for half an hour with a tutor, on top of the subscription.

Worth knowing before you budget, and I couldn't find it stated on a single page ranking for this query.

The Duolingo price I can't honestly quote

Here's where I have to do something the other pages don't, which is decline to give you a number.

Duolingo's US App Store listing shows Super Duolingo in-app purchases at $9.99, $12.99, $83.99, $95.99 and $119.99. The store attaches no billing term to any of them. There's a 14-day free trial, and the perks are "no ads" plus "fun perks like Unlimited Energy & Monthly Streak Repair."

Which of those is the monthly price and which is the annual? The listing doesn't say. I tried duolingo.com/plans and it renders its prices in the browser rather than in the HTML, so an automated read comes back empty.

So the accurate statement is a range: Super Duolingo's App Store prices run from $9.99 to $119.99 depending on term and plan. Anything more precise than that is a guess dressed as a fact.

Compare what the ranking pages do with that gap. One says Super is "$13 per month" monthly and "$7 per month" annual. One says "$6.99/mo" under the name Duolingo Plus, a brand Duolingo retired. One says "about $7 a month." Three confident figures, three different numbers, zero sources.

I can't confirm a price for Duolingo Max either. It's widely reported at $29.99 a month, and I couldn't find that on a Duolingo-owned page, so I'm not printing it as fact. Same for the family plan.

The practical advice is duller and more useful than a number: open the app and look at your own purchase screen. Duolingo prices vary by region and discounts the annual term heavily, so the figure on any comparison page was going to be wrong for you anyway.

And the number that matters most here is one everybody agrees on. Duolingo's free tier is real. The full core course, ad-supported, gated by energy. Paying removes ads and the energy cap; it doesn't buy you more teaching. Be clear about what the upgrade is actually for before you make it.

A year of each, in money

Monthly fees are the wrong unit for this decision. Here's the annual version, which is what you'll actually spend.

The row that should stop you is the last one. The $399 tutoring package buys 10 half-hour sessions, so 5 hours, at about $80 an hour. Booked on an independent tutor marketplace the same $399 goes much further: on italki's own Spanish directory, 20 teachers on one page had a median standing rate of $19 an hour when that page was sampled on 23 August 2026, which puts $399 nearer 21 hours than 5.

Now the honest part, because that comparison flatters neither product. Neither app gets you to conversational fluency on its own, and neither company claims it does. The State Department's own Foreign Service Institute puts Spanish at 24 to 30 weeks and 552 to 690 class hours to reach professional working proficiency, on a week of 23 class hours plus 17 hours of self-study. No $159 subscription is buying you that. A subscription buys the vocabulary and structure that makes the hours you do spend go further, which is a real thing to buy and a smaller one than the marketing suggests.

Two methods, and which one suits you

The products teach in genuinely different ways, and this is the part worth deciding on.

Rosetta Stone runs on immersion with no English in it. Its own description of Dynamic Immersion: the method "taps into your brain's innate ability to learn, no translations or rote memorization needed," and in Chat Missions "the entire conversation is in your target language." You work out what el perro corre means from a picture of a dog running, rather than being told.

That feels confusing for about a week, and then it works for a lot of people, because you never build the habit of translating in your head first. It also means that when you want to know why the verb changed, the app stays quiet.

Pronunciation is where Rosetta Stone is genuinely strong. TruAccent, per Rosetta Stone, "compares your voice to data from millions of native speakers and gives you immediate feedback on your pronunciation." Duolingo has speaking exercises, but they put the target sentence on screen before you say it, so the retrieval work is done for you.

The new Sapphire pieces are worth naming because no ranking page has them yet. Chat Missions are scenario conversations with an objective (ask for water at a restaurant, negotiate with a market vendor) and real-time correction on spelling, grammar and vocabulary. Sapphire Studio generates your own materials: a reading guide, an adapted text, a vocabulary list, flashcards, a matching game or an email draft, and you can edit the generated text or add items with a plus icon.

Duolingo runs on translation and habit. You translate sentences, the streak pulls you back tomorrow, and the notifications are shameless about it. People sneer at the streak, and meanwhile it's the reason plenty of learners open a language app 400 days running.

Getting somebody to show up daily is the hardest problem in this category, and Duolingo solved it better than anyone has.

Two Duolingo facts the ranking pages get wrong, both easy to check. Its spaced repetition is real: Settles and Meeder published a trainable scheduler, half-life regression, at ACL 2016, peer-reviewed and open-sourced. And the "Duolingo only reaches A2" line expired on 22 April 2026, when Duolingo announced nine courses including Spanish now carry content through CEFR B2, at a Duolingo Score of 129. Read that precisely: it's a claim about how far the course content reaches, not about where learners land.

Try both for free first

You can answer the immersion-versus-translation question yourself this week, for nothing, and it'll beat any review including this one.

Duolingo's free tier is permanent, and it's the whole core course. Work through a week of it.

Rosetta Stone's 7-day trial runs alongside it. Do the same unit in both, the one you'd actually need first, and notice which one you're avoiding on day 4. That avoidance is the real signal, and it shows up faster than progress does.

Then cancel whichever you don't want during the trial, because of the auto-convert to monthly. And if you buy direct and change your mind, the 14-day satisfaction guarantee is your backstop.

What the evidence actually shows

Both companies cite research. It's worth knowing where it comes from, because I went and read it.

Rosetta Stone and Duolingo have both commissioned efficacy studies from the same research pair, Roumen Vesselinov and John Grego, who between them produced 16 such reports for 10 different app companies over 16 years. Rosetta Stone commissioned studies in 2009 and again in 2019; Duolingo's is from 2012.

Credit where it's due: 15 of those 16 reports name their funder, usually on page 5 under study limitations. That's a better conflict statement than plenty of peer-reviewed work carries.

The problem is the league table built on top of them. The published ranking puts Rosetta Stone's 2019 study at 13 study hours to cover a first college semester and Duolingo's 2012 study at 34, bottom of the table. Every row in that table comes from a study paid for by the company in that row, and the table has no column saying so.

A 2021 paper in Foreign Language Annals, four of whose five authors list Duolingo as their affiliation, says the reports "were published on company websites as white papers, not peer-reviewed journal articles," and that because pretest scores differ, "it is hard to compare the effectiveness across these products." A peer-reviewed journal saying the league table can't carry the weight, with Duolingo's own staff as authors.

Duolingo's 34 hours also comes from a 2012 study of a product that no longer exists. I wrote the full audit up in one lab, fourteen apps. The short version: treat both companies' efficacy claims as marketing with footnotes, and don't let a 13-against-34 comparison decide anything.

Side by side

Which one to buy

Buy neither, for now, if you're at a standing start and unsure. Duolingo's free tier costs nothing and answers the "will I stick with this" question better than a paid month of anything.

Buy Duolingo Super if the free tier has already earned its place in your day and the ads and energy cap are what's annoying you. You're buying the removal of friction on a course you already use, which is a clean thing to pay for.

Buy Rosetta Stone if pronunciation is your sticking point, or if you want someone to hand you a structured path and you'd rather work a language out from pictures than translate your way through it. The $159 annual is the sensible tier; the monthly costs $80.88 more over a year for flexibility you probably won't use. And if you're learning several languages, all 25 are included, which changes the arithmetic entirely.

Take the tutoring add-on only if booking a tutor yourself is the thing stopping you. It's convenience at roughly triple the going rate.

What both of them can't do

One limit is shared, and it's structural rather than a flaw either company could patch.

Both teach a fixed word list in a fixed order. Rosetta Stone's support page describes a curriculum "divided into units, lessons, and activities" with units covering "big topics like 'Work and school', 'Home and health'." Duolingo's tree works the same way. Neither can teach you the word you reached for in a real conversation on Tuesday and didn't have.

That word is the one you're most likely to need again, because you already tried to use it. And it's the one no closed curriculum will ever contain.

Verbamor is built for that gap. You record or type what came up in a tutor lesson, and it builds fill-in-the-blank cards from the words you actually fumbled, with audio and an image, scheduled by FSRS. Five languages, Spanish, French, Italian, Portuguese and Russian. It's $19.99 a month or $129.99 a year, iOS only, with a 7-day trial on the annual plan and no free tier. Both are US prices, read off the app's own plan definitions on 21 September 2026; the App Store sets a different number in other countries.

It teaches you nothing new, which is the honest limit. It holds what a teacher already gave you. If nobody's giving you material, there's nothing for it to hold, and one of the two apps above is the better purchase.

Questions people actually ask

---

Sources

Every load-bearing claim Verbamor makes is traced to its paper on the research page.