One concept guide
$2.99 one time
A short visual explanation, guided practice, and a scored check.
NovaSprout Clear · Your AI tutor
Watch a short animation, then create a lesson on your own topic.
Press play to try it
This short fraction preview combines illustrated animation with AI-generated narration. Watch, pause, replay, then try a question.
Illustrated animation preview · 34 seconds · suitable for learning halves and quarters
Start here
Preview a real concept guide free, or use approved access to clarify the student's own topic.
One concept guide is $2.99. Monthly NovaSprout Clear is $14.99/month for up to 20 guides per billing month. Each guide includes a visual explanation, guided practice, a quick quiz, a visual PDF, and up to 10 follow-up questions within 24 hours of generation.
Monthly access renews automatically until cancelled. Cancel in Manage billing; access continues through the paid period. Unused monthly lessons do not roll over. AI can make mistakes; review important answers with a parent or tutor.
$2.99 one time
A short visual explanation, guided practice, and a scored check.
$14.99 / month
Up to 20 concept guides per billing month, with follow-up questions.
This restores website purchases. App Store purchases remain in the iOS app. If you lost the code, contact NovaSprout with your Stripe receipt.
Need human help?
The AI tutor works independently. If you want a person to help with a persistent gap or learning plan, you can request human tutoring separately.
The learning approach
Inspired by Benjamin Bloom’s mastery-learning research: find the gap, work on it, and check what the learner can do independently.
Begin with a small prerequisite question and an attempt.
Use a clear explanation, a worked example, and a targeted hint.
Correct missed quiz answers. Fresh questions help distinguish understanding from remembering an answer.
Save the lesson, revisit it tomorrow without notes, and bring persistent gaps to a human tutor.
Bloom’s 1984 paper reported an average advantage of about two standard deviations for tutoring combined with formative checks and corrective work in the studies described. That historical result is not a measured NovaSprout outcome or a guarantee for every learner.
We use it as a design challenge. First-try answers, corrections, and later independent performance are different signals. NovaSprout has not demonstrated a two-sigma gain; that would require a controlled evaluation of unaided learning and retention.
Read Bloom’s original paper (1984) · Read research on AI tutoring safeguards (2025)