Both platforms serve UKMLA CPSA candidates with AI simulation inside a broader learning ecosystem, and they differ in scope and evidence in ways worth stating precisely rather than blurring.
Quesmed OSCE-AI, on its own terms
An integrated undergraduate ecosystem combining written questions, teaching resources, more than 450 OSCE stations with more than 340 AI-enabled, data-interpretation stations, group study, mobile apps and unlimited voice and text attempts without a credit system. Quesmed has also published a small examiner-calibration pilot, ten transcripts marked by four doctors and its AI examiner, with the AI within one global band of a human examiner 88% of the time against 87% human-to-human agreement, genuinely rare transparency in this category, and a small, company-run pilot rather than independent validation.
iatroX, on its own terms
Eighteen examination-specific simulation tracks including a UKMLA CPSA track organised around the MLA content map's actual domains, connected to the AKT-focused question bank and Socratic Tutor in the same account, with full mock circuits, transcript-linked feedback and prescribed remediation. Coverage extends beyond undergraduate assessment into postgraduate and international examinations, meaning the same subscription continues past finals.
The comparison
Scope: Quesmed is undergraduate-and-MLA focused; iatroX spans undergraduate through postgraduate and across jurisdictions. Simulation model: Quesmed offers unlimited attempts on a fixed station library; iatroX offers exam-specific tracks with prescribed next cases and full circuits. Data interpretation: Quesmed's dedicated data-interpretation stations are a specific strength; iatroX's CPSA track includes interpretation-domain cases within its content-map organisation. Published marking evidence: Quesmed has published a pilot; iatroX describes its clinician-review methodology and continuous-improvement process publicly, and this comparison does not claim iatroX has published equivalent examiner-calibration data. Integration with written preparation: both connect simulation to their own question content; iatroX's Tutor-led remediation loop is its specific differentiator.
Verdict
A candidate whose preparation is bounded to UK undergraduate finals and the MLA, who values unlimited attempts and Quesmed's integrated undergraduate depth, is well served by Quesmed. A candidate who wants simulation connected to Tutor remediation, organised by MLA domain with a prescribed next case, and a subscription that continues into postgraduate training, is the candidate iatroX's track is built for. Both platforms' automated marking should be treated as formative rather than as a readiness verdict, and both should be supplemented with your own medical school's specific CPSA format information, since the CPSA is locally delivered rather than nationally standardised.
Choosing based on what you already use
Many UKMLA candidates already run a Quesmed subscription for its written question bank and teaching resources before OSCE-AI ever enters the decision, in which case the marginal cost of also using its included simulation is effectively zero, a genuine point in its favour regardless of how it compares feature by feature. A candidate starting from nothing, or already using iatroX for other UK or international exams, faces a cleaner comparison on the dimensions above. Confirm current pricing and feature scope for both directly before deciding, since this category updates frequently enough that any specific figure here should be treated as a snapshot rather than a permanent fact.
The transparency point worth weighing on its own terms
Quesmed's published marking pilot deserves to be weighed as a genuine point in its favour independent of everything else in this comparison, because publishing any calibration data at all, even from a small sample, is rarer in this category than it should be. A candidate who values evidence over marketing claims should factor that transparency into the decision directly, while still applying the same scrutiny this cluster's dedicated critique of that specific pilot recommends, ten transcripts is a genuinely useful early signal and not, on its own, proof of examiner-equivalent marking at scale.
The strongest signal for choosing between them is often the simplest: try both platforms' free or trial access directly on your own examination content before committing, since a feature comparison on paper rarely captures how a specific candidate actually experiences a platform's interface, pacing and feedback style in practice. Whichever you choose, the discipline of reviewing your own transcripts honestly, rather than trusting either platform's score alone, is what ultimately determines how much genuine value you extract from it.
Frequently asked questions
Does iatroX have a published marking-calibration pilot like Quesmed's?
This comparison does not claim so: Quesmed's published pilot is a genuine and rare transparency point, and iatroX's published material describes its clinician-review and improvement methodology rather than an equivalent examiner-agreement study.
Which is better for data-interpretation stations?
Quesmed's dedicated data-interpretation stations are a specific strength worth weighing directly; iatroX includes interpretation cases within its content-map-organised CPSA track.
Does either platform reproduce my medical school's exact CPSA format?
No: the CPSA is designed and delivered locally by each school under GMC quality assurance, so both platforms cover the nationally consistent content-map domains, which you should supplement with your own school's specific station format.