AMP vs Synthesis Tutor
Synthesis Tutor is a highly rated app for children 5 to 11, with a spoken tutor, on-screen manipulatives and read-aloud for pre-readers1,6,9. It is built for kindergarten through grade 5, states no mastery threshold, and we found no scheduled review on its pages1,2. AMP runs from counting through precalculus, picks every problem from the learner's answers, and brings each learned skill back on a fixed schedule.
Why AMP
It keeps going after grade 5
Synthesis Tutor is designed for ages 5 to 11 and covers the standard K–5 curriculum1. Its listed topics run from counting and cardinality to fractions and algebraic thinking1. AMP runs from counting through precalculus in one program, 336 skills on one map. A learner who races ahead, or finishes grade 5, keeps the same record and keeps going.
Learned skills come back on a schedule
AMP brings each learned skill back after 6 hours, 1 day, 3 days, 1 week, 2 weeks and a month, mixed with other topics, and a miss restarts the schedule. We found no scheduled review of earlier lessons on Synthesis's pages1,2,3. In a randomized trial with 787 seventh graders, mixed and spaced practice scored 62% on the researchers' test a month later, against 38% for the same problems practiced in blocks14.
The research: Spaced, mixed review
A mastery rule you can read
Synthesis says a child moves on once they show they have understood and mastered the material, and gives no threshold1. In AMP a skill is Strong when four of the last five answers are right without help. An answer given after a hint or the lesson is recorded but does not count. Two misses in a row send practice to what the skill builds on.
The research: Mastery, and repair of gaps · The research: One-step hints
Every problem chosen from the learner's answers
Synthesis Tutor adapts inside a lesson and moves on by lessons. On its Starter plan the child chooses lessons from a map; placement checks and a lesson planner come with Pro1. AMP does this for every learner: it places the learner within the first 24 problems and keeps a level in each of 14 strands, then picks each next problem as new work, a prerequisite, or review.
The research: Practice at the learner’s level
The method on the page, every time
Synthesis starts each topic with hands-on action on screen and moves to symbols after6. Its page for schools promises step-by-step guidance when students make mistakes3. AMP gives every problem type a written lesson with a worked example, and after every answer shows the correct answer and the worked steps. Each of its 961 reference lessons includes a guide for the parent. Across 55 studies, worked examples improved math performance by g = 0.4815.
The research: Lessons with worked examples · The research: Feedback after every answer
Paper when the screen goes off
We found no workbooks or printables on Synthesis's pages, and it recommends an internet connection of 50 Mbps or faster1,2,3. AMP makes unlimited printable workbooks for each learner, each with a worked answer key. Results entered from paper count toward the learner's level.
Side by side
| AMP | Synthesis Tutor | |
|---|---|---|
| Price | $599 a year or $59 a month for the first child, then $399 or $39 for each more. Two parents practice free. Full refund within 30 days. | Starter $119 a year for up to seven children; Pro $239 a year; Live, with six tutor sessions, $480 a month1,2. Fees not refunded except where law requires4. |
| Who can buy | Families, tutors and schools, directly. Free to start without a card: 100 answers and 5 workbooks. Arizona ESA, Arkansas EFA and New Hampshire EFA funds pay for it. | Families online, with a 7-day free trial and no card1. Schools through a sales contact, with no price published3. Classrooms also get it as Dojo Math in ClassDojo8. |
| Range | Counting through precalculus in one program: 336 skills on one map. | Ages 5 to 11: the standard K–5 curriculum, from counting to fractions and algebraic thinking1. |
| Finding the starting point | No separate test. Placement takes the first 24 problems of practice. After that, every answer adjusts a level in each of 14 strands of math. | Placement checks on the Pro plan. On Starter, the child picks lessons from a map1. |
| Teaching | A one-step hint and a written lesson with a worked example for every problem type. Worked steps after every answer. Narrated lesson films. | Lessons of 5 to 6 minutes, guided by a tutor with a voice and built on on-screen manipulatives1,6,7. Read aloud automatically for children under 71. |
| What decides the next problem | AMP chooses every problem from the learner’s answers: new work just within reach, a prerequisite, or review. | On Pro, a lesson planner picks the next lesson; on Starter, the child chooses. Inside a lesson, the software adapts to answers1. |
| Mastery | Four of the last five answers right without help. An answer after a hint doesn’t count. Two misses in a row send practice back to what the skill builds on. | The child moves on after showing understanding. No threshold is published1. |
| Review | Scheduled. Learned skills come back after six hours, a day, three days, a week, two weeks and a month, mixed with other topics. | No scheduled review of earlier lessons found in its published materials1,2,3. |
| Unlimited workbooks made for each learner, each with a worked answer key. Results entered from paper count. | No workbooks or printables found in its published materials1,2,3. Online only, with 50 Mbps internet recommended2. | |
| What the adult has to do | Nothing to plan, teach or grade. AMP places the learner, chooses each problem, checks it and schedules review. | Little. The maker calls it hands-off for parents and teachers. Adults set up accounts and read reports1. |
| Reports | The level in every strand with its likely range, skills, time, and the actual problems behind each number. For teachers, who needs them today. | Progress reports. Pro adds session summaries and weekly progress emails1. No gradebook found in its materials for schools3. |
| Evidence | A published rationale (ESSA Tier 4) with the research behind each part, public source code, and outcome studies from 2027, each registered before it starts. | No outcome study found on its site, in Evidence for ESSA or in the What Works Clearinghouse5,12,13. |
What Synthesis Tutor is
From Synthesis
Synthesis Tutor is made by Synthesis, a California company founded in 2020 by Josh Dahn and Chrisman Frank. It grew out of a class at an experimental school started on the SpaceX campus in 20142. The tutor is for ages 5 to 11 and covers kindergarten through grade 5 math1. It runs on iPad, Android tablets and in a browser, and its App Store listing shows 4.6 stars from about 10,000 ratings2,9. Classrooms meet a version of it as Dojo Math inside ClassDojo8.
A lesson runs 5 to 6 minutes and teaches one idea7. A tutor with a voice offers guidance on every problem1 while the child acts on the screen: smashing a square with a hammer to see fractions, or firing a blaster to build multiplication6. Short checks inside each lesson decide when the child moves on1. Adaptive flashcards and games sit beside the lessons9. For children under 7, every word and answer choice is read aloud1.
The maker's FAQ says Synthesis Tutor is not ChatGPT1. Its children's privacy policy gives, as examples of its AI language services, help pronouncing a child's first name and choosing grammatical gender when a lesson is translated to Spanish8. This page compares it as a lesson program, not a chatbot.
What Synthesis Tutor does well
- Made for young children. Everything is read aloud for children under 7, the voice speed adjusts, and children can answer by speaking, in beta1,9. AMP has no read-aloud of problems.
- Each topic starts with hands-on action on screen before symbols6. AMP has ten built-model problem types, not a manipulative at the heart of every lesson.
- A lower price, most of all for big families: $119 a year covers up to seven children2, after a 7-day trial with no card1.
- Apps for iPad and Android tablets, lessons in Spanish, games, and flashcards that track speed as well as accuracy1,9,10. AMP has none of these.
- A human tutor when wanted: the Live plan, at $480 a month, includes six live sessions with a human tutor1. AMP has no live tutors.
The evidence
Synthesis Tutor
We found no outcome study of Synthesis Tutor on its own site. Synthesis Tutor has no entry in Evidence for ESSA or the What Works Clearinghouse (checked November 2026)12,13.
Synthesis's blog post on whether Synthesis Tutor gets results reports no data from Synthesis Tutor5. It describes the DARPA Digital Tutor and says Synthesis partnered with the team that built it5. A 2020 summary that Synthesis links describes that program as a tutor built around 2009 to train new Navy recruits as IT workers11. The summary gives no number of students in each group and does not say how they were assigned11. It was a different product, for adults, in a different subject.
The page for schools calls the product "the world's first superhuman math tutor" and states that 98% of teachers reported improved math skills, with no source, sample or date given3. The pattern is the simplest one: no outcome study we could find.
AMP
We built AMP's rules on findings that have held up in trials. In a randomized trial with 787 seventh graders that meets What Works Clearinghouse standards without reservations, mixed and spaced practice scored 62% a month later, against 38% for blocked practice14. Across 40 studies of computer-based feedback, feedback that explains averaged an effect of 0.49, against 0.32 for the correct answer alone and 0.05 for right or wrong only16. AMP applies both at every problem. We expect that to show in outcomes; that is our bet, and we have committed to testing it in public.
AMP is new, and its own outcome studies have not run yet. They begin in 2027: growth on a national test against hours of use, then comparison groups, then randomized trials. Each will be registered before it starts and published whatever it finds. Until then, the research behind each part of AMP is laid out in our report, and the code is public.
Choose Synthesis Tutor if
- Your child is 5 to 8, or not yet reading, and needs every problem read aloud.
- You want lessons in Spanish, or an app on an iPad or Android tablet.
- You have several young children and a tight budget. One $119 plan covers up to seven.
- You want math games and flashcards that track speed, or an occasional live tutor from the same company.
Choose AMP if
- Your child is ahead, or in grade 4 or 5, and will need math past grade 5.
- You want a stated mastery rule that help cannot satisfy, and old skills brought back on a schedule.
- You want to see the method: a worked example for every problem type and worked steps after every answer.
- You want paper too: workbooks made for your child, with answer keys, whose results count.
- You teach children of different ages and want one program from counting to precalculus.
Questions
Is Synthesis Tutor an AI chatbot?
Its maker says it is not ChatGPT1. Its children's privacy policy gives help pronouncing a child's first name and choosing grammatical gender in Spanish translations as examples of its AI language services8. AMP runs no AI model while a learner works; every problem, hint and explanation comes from tested code.
Does Synthesis Tutor cost less than AMP?
Yes. Starter costs $119 a year and Pro $239 a year1. The About page gives $29 a month or $119 a year, with one plan covering up to seven children2. AMP costs $599 a year or $59 a month for the first child, then $399 or $39 for each more, and covers counting through precalculus, a level in every strand, scheduled review and unlimited printable workbooks.
Is Synthesis Tutor based on DARPA research?
Synthesis says it partnered with the team behind the DARPA Digital Tutor5, which a summary Synthesis links describes as training new Navy recruits as IT workers11. We found no study of Synthesis Tutor itself on its site, in Evidence for ESSA or in the What Works Clearinghouse12,13.
What should my child use after grade 5?
Synthesis Tutor is designed for ages 5 to 111. AMP continues through Algebra 1, Geometry, Algebra 2 and Precalculus, with the same record of every skill.
Can we use Synthesis Tutor and AMP together?
Yes. They do different jobs: Synthesis Tutor introduces K–5 ideas with on-screen action6, and AMP finds the level, schedules review and prints workbooks. AMP's first 100 answers are free, with no card.
Sources
- Synthesis Tutor (Synthesis)
- About Synthesis (Synthesis)
- Synthesis Tutor for educators (Synthesis)
- Terms of Service (Synthesis)
- Does Synthesis Tutor get results? (Synthesis blog)
- The science of Synthesis Tutor 1: the embodiment of mathematical knowledge (Synthesis blog)
- Why math finally clicks with Synthesis Tutor (Synthesis blog)
- COPPA policy (Synthesis)
- Synthesis Tutor on the App Store
- Synthesis Math Tutor on Google Play
- DARPA Digital Tutor: four months to total technical expertise? (LessWrong, 2020)
- Math programs (Evidence for ESSA)
- Products (What Works Clearinghouse)
- Rohrer, Dedrick, Hartwig & Cheung (2020), A randomized controlled trial of interleaved mathematics practice (What Works Clearinghouse review)
- Barbieri et al. (2023), A meta-analysis of the worked examples effect on mathematics performance
- Van der Kleij, Feskens & Eggen (2015), Effects of feedback in a computer-based learning environment, Review of Educational Research 85(4)
Checked November 2026.
Synthesis Tutor is a trademark of its owner. AMP is not affiliated with it. Found an error? Write to skylar@corpora.inc.
More comparisons
- AMP vs ChatGPT: A chatbot answers what a child asks; AMP chooses each problem, checks every answer and schedules review.
- AMP vs Photomath: Photomath solves the problem a learner photographs. AMP chooses the next problem and asks the learner to solve it.