Your Codex data, at a glance.
Three canonical families under an active commercial license — normalized to the APRS standard, snapshotted monthly, ready to join.
Licensed record volume
Provenance feed
LiveHow Codex data flows
Licenses
Your commercial keys and signed download endpoints. Keys are scoped per dataset and rotate with each snapshot.
Add a dataset
5 available · $299 / dataset / monthCompare tiers
you are on Commercial- 100K-row stratified sample per family
- Parquet via Hugging Face
- Schema spec access
- Full datasets
- Monthly immutable snapshots
- Entity graph export
- Full dataset, all records
- Parquet · CSV · JSON Lines · GeoParquet
- Monthly immutable snapshots
- Signed R2 download URLs
- Schema changelog + email support
- Entity graph export
- Everything in Commercial
- Full entity graph export (8–12M edges)
- All formats + Markdown-KV llm_text
- Hosted SQL query API · webhook push
- Weekly update cadence
- Custom schema extensions
- Dedicated support + SLA
Data catalog
Every family ships the same APRS envelope, pre-computed AI labels, and shared join keys — so any two datasets join without wrangling. Select a family to inspect its schema and sample records.
Downloads
Each snapshot is content-addressed and delivered through a signed Cloudflare R2 URL bound to your license. URLs rotate monthly and expire — regenerate from this page any time.
Integrate
Copy-paste starters for your licensed families. Snapshots are plain Parquet (GeoParquet for the grid) — no SDK, no import job, no API key. Signed URLs expire; regenerate them from Downloads.
Entity graph
Entities and relationships resolved across families — places, parcels, vessels, organizations, permits. This preview shows a synthetic sample; the Enterprise export carries the full 8–12M-edge graph. Drag to pan, scroll to zoom.