Compare

Where we win, and where we don’t.

An honest comparison against the platforms you are already using. It includes the rows we lose — a comparison that only flatters its author is one nobody finishes reading. Assessed August 2026.

For the learner

A queue that explains itself, a prediction that reports its own error, and a record that survives graduation.

For the faculty

Authoring that takes fifteen minutes instead of four hours, with psychometrics on your own items rather than someone else’s.

For the institution

Accreditation output as a by-product of ordinary learning, and the option to run all of it inside your own tenant.

Feature by feature

✓ available · ◐ partial or configuration-dependent · — not available. Where a competitor is genuinely better, the cell says so.

CapabilityVeloLibraryUWorldAMBOSSOsmosisAnkiTrueLearnHealthStream
Core learning
Adaptive board-style QBank
Bank size and brand recognitionGrowingBest in classCommunity
Spaced-repetition flashcardsBest in class
Video lecture libraryPhasedBest in class
Clinical case simulation with an AI tutor
Detailed rationale per itemCommunity
Offline practice with conflict-free sync
AI & personalisation
AI-generated personal study plan
Explainable item selection
Conversational clinical-case tutor
AI-assisted item authoring for faculty
Calibrated score prediction with error
Inspectable AI run ledger
Institution & faculty
Faculty authoring on the institution’s own bankCommunity
Psychometrics on institution-authored items
Cohort analytics for educators
Curriculum mapping and coverage
LMS / SIS integration (LTI 1.3, OneRoster)
Accreditation
ACGME milestone tracking
CCC evidence packet export
ACCME-ready CME issuanceV2
Immutable audit trail
Deployment & commercial
Self-hosted / data-sovereign optionLocal files
Institutional site licenceFree
Priced per learner per period
One record across school → residency → CME

The landscape, fairly described

Each of these is genuinely good at something. Knowing which one is right for you matters more to us than winning every row.

VeloLibraryInstitution-native medical learning

One learner record from first lecture to last CME credit, with the authoring engine, psychometrics and accreditation output an institution needs. Newest bank in the category — we compete on the loop and the institutional layer, not on catalogue size.

UWorldThe benchmark QBank

The best-known bank in medicine, with explanation quality that set the category standard. Individually priced, closed to institutional authoring, and it ends at the exam.

AMBOSSQBank + integrated library

Excellent cross-linking between questions and a genuine reference library, and strong in-hospital use. Institutional analytics and accreditation tooling are not its focus.

OsmosisVideo-first learning

The best visual explanations in the market and a strong student brand. Lighter as an assessment engine, and not built for program-level reporting.

AnkiFree spaced repetition

Unbeatable retention mechanics and a vast community deck ecosystem, for free. Setup is the barrier, nothing is institution-owned, and there is no analytics or accreditation surface at all.

TrueLearnInstitutional QBank

Sells to programs rather than individuals and reports at the cohort level, which is the closest commercial posture to ours. Lighter on AI, authoring and the longitudinal record.

HealthStreamHealthcare compliance LMS

The system of record for hospital compliance training and CE, with real CME plumbing. It manages completion rather than teaching clinical reasoning.

MedHub / New InnovationsResidency management

The GME office’s system of record for evaluations, duty hours and CCC reporting. Their milestone data is manually entered opinion; ours is generated from learning evidence. V1 connects rather than competes.

Why programs start looking

Not feature gaps. Structural ones — the things that cannot be fixed by adding a screen to what you already have.

The record dies at graduation — four years of calibrated performance data becomes a transcript line.

Program director interviews; the residency starts from zero

Institutional banks stay small because authoring starts from a blank page and faculty time is the constraint.

The reason most institution-owned banks stall under 300 items

Milestone evidence is transcribed by hand every cycle, the week before the CCC meets.

ACGME reporting workflows in current GME tooling

Per-learner pricing punishes exactly the institutions with the most learners.

Every incumbent prices per learner per subscription period

Cloud-only architecture ends the conversation at the CISO desk.

Public medical schools and health systems increasingly require data sovereignty

Adaptivity is a black box, so faculty cannot defend why a learner saw what they saw.

No incumbent exposes its selection rationale