The database developer
One SQL statement, same tables, same permissions. Nothing to deploy.
PQL adds TRAIN MODEL and PREDICT to DuckDB. Your agents answer “what happens next?” in the SQL they already speak. No pipeline, no export, no second system.
INSTALL pql FROM community; LOAD pql;TRAIN MODEL lapse
PREDICT EXISTS(transactions)
FOR customers AT last_seen_at
HORIZON 30 DAYS;
-- seconds later: score, baseline
PREDICT EXISTS(transactions)
FOR customers USING MODEL lapse;One SQL statement, same tables, same permissions. Nothing to deploy.
A first defensible model in seconds, not weeks. Leakage handled by construction.
Trains, scores and predicts on its own, in the SQL it already writes. No Python and no training stack to hand it.
SaaS renewals, e-commerce reorders, missed payments, support tickets: the shape is always the same.
TRAIN MODEL lapse
PREDICT EXISTS(transactions)
FOR customers AT last_seen_at
HORIZON 30 DAYS;TRAIN MODEL risk
PREDICT EXISTS(scan_events
WHERE scan_events.type = 'exception')
FOR shipments AT dispatched_at
HORIZON 3 DAYS;PQL splits time at each row's AT value. Features come only from before it, and the engine counts the label after it.
RelBench is the Stanford benchmark for relational prediction.
| rel-f1 task | PQL | Gradient boostingBoostinghand-built features | Graph neural netGNNRelBench RDL, GPU |
|---|---|---|---|
| Qualify top-3 next month?AUROC | 0.809 | 0.842 | 0.809 |
| Fail to finish a race?AUROC | 0.707 | 0.701 | 0.741 |
| Finishing position?MAE, lower is better | 4.123 | 4.403 | 4.259 |
| Hardware, training time | Laptop CPU 3 to 15 s | Laptop CPU about 1 s | NVIDIA L4 GPU 17 to 78 s |
One win, one tie, one loss against each, without moving a row.
Load the extension in any DuckDB shell and train your first model on the tables you already have.
INSTALL pql FROM community; LOAD pql;