Pricing

Query pre-computed, deterministic signals — changepoints, HMM regime states, z-scores — from 60+ integrated public data sources. No API integration, no feature-engineering pipeline to maintain.

• Daily-refreshed data across 47 sources — SEC EDGAR, CFTC, GDELT, AIS, and more
• Deterministic feature extraction (BOCPD, HMM, spectral) — no LLM, fully auditable
• Query by source, time range, and row limit; a source catalog endpoint lists everything queryable
• Higher tiers: custom universe, batch delivery, dedicated support

A deterministic, no-LLM pipeline that extracts entities and relationships from ingested documents — PDF, news, CSV, plain text — with every mention and link tracing back to the exact source sentence it came from. Cross-document co-occurrence — the same pair recurring across sources, not just one — is the signal.

• Entity search — every mention and relationship cites its source document and sentence; confidence scores are deterministic (co-occurrence-based), not model output
• Cross-document co-occurrence analytics — surfaces entities and relationships that recur across the ingested corpus
• Graph export (adjacency list) and degree-centrality / hub analysis
• Enterprise: custom entity types, private document ingestion, on-prem option

Read-only visibility into the DAG-based execution engine that runs our own pipeline in production — run history, status, and timing per DAG.

• List recent runs by DAG name, with status and timing
• Inspect a specific run's node-level results

BETA Submitting your own custom DAGs for execution is in closed beta — contact us if you want in.