The model scores well. Is the report going to survive a reviewer?

Paste the methods and results of a clinical prediction model, and the table of what it scored on each split. Before you sign in, this page reads the report properly: it counts the TRIPOD+AI items you have actually answered, divides your events by your candidate predictors, works out what predicting no event for everyone would have scored, screens every predictor against your own stated time of prediction, and checks each metrics row's numbers against each other. Then four lanes work the report itself.

Both examples ship with a saved model run for every lane, so you can see all four outputs end to end without signing in and without spending a credit.

The model report nothing pasted yet

Labelled lines, prose, or both. Drag a file in, or Read in your browser. Nothing uploads until you run a lane.

Performance table — optional, but it turns on the interval, Bayes-rule and Brier checks no metrics table

CSV, TSV, semicolon or markdown pipes. Drag a file in, or Columns read: split, n, events, prevalence, auc, auc_lo, auc_hi, brier, cal_slope, cal_intercept, oe, sens, spec, ppv, npv, threshold.

Where this model is

Who is going to read this

Anything the report does not say — reviewer comments, the analysis you have not written up yet, the deadline

Paste a report to price the run.

The model's raw reply

The structured parse failed twice, so here is the reply verbatim rather than nothing.

The model scores well. Is the report going to survive a reviewer?

Instant read — free, in your browser, no model

Where the model and the browser disagree

Findings

Every free-read flag, accounted for

What you told it, checked against the report

Not assessable from what was supplied

Summary

The model's raw reply

Your past reviews

How this was produced