Skip to main content
Predexy distinguishes between two related but distinct concepts: platform markets and canonical questions. Understanding the difference helps you choose the right endpoint for your use case and interpret the data you receive correctly.

​
Platform markets vs canonical questions

A platform market is a single listing on a specific prediction market venue — for example, Polymarket’s “Bitcoin to $100,000 before 2026?” or Manifold’s “Will BTC hit six figures by end of 2025?”. Each platform market has its own title, volume, liquidity, and outcome prices, and it belongs to exactly one platform. A canonical question is Predexy’s aggregated representation of the same real-world question across all platforms that list it. When the semantic matching pipeline links two or more platform markets together, they share a canonical question. The canonical question carries a volume-weighted index price, a divergence score, and a market_count that tells you how many platforms list it. A market_count of 1 means only one platform covers that question; a market_count of 3 or more means you have cross-platform pricing data, which is where arbitrage opportunities emerge. Use GET /api/v1/discover to browse canonical questions. Use GET /api/v1/questions/ to drill into the platform markets attached to a specific canonical question.

​
Markets endpoints

The /api/v1/markets endpoints give you direct access to the underlying platform market layer, independent of canonical question groupings. All market data is populated by the ingestion pipeline, which runs continuously — so what you receive is always fresh from the last scan cycle.

​
List markets

GET https://api.predexy.com/api/v1/markets
Returns a paginated list of platform markets across all venues. Supports filtering, sorting, full-text search, and cursor-based pagination. Use this endpoint when you want to browse raw market data rather than the canonical question layer. Supported query parameters:
ParameterTypeDefaultDescription
sortstringvolume_24hSort by volume_24h, liquidity, newest, or ending_soon.
windowstring24hTime window for volume metrics: 24h, 7d, 30d, or all.
platformstring—Filter by platform slug, e.g. polymarket.
categorystring—Filter by category.
qstring—Full-text search query.
cursorstring—Opaque pagination cursor from the previous response’s meta.next_cursor.
limitnumber50Maximum items to return (1–200).

​
Get a single market

GET https://api.predexy.com/api/v1/markets/{id}
Returns full details for a single platform market by its UUID, including all outcomes and current prices. 
GET https://api.predexy.com/api/v1/markets/trending
Returns the top markets ranked by 24-hour volume. Pass limit to control how many results you receive (1–200, default 50). Use this to power a “hot markets” widget or to surface the most actively traded questions.

​
Sparklines

GET https://api.predexy.com/api/v1/markets/sparklines?ids=<uuid>,<uuid>,...
Returns compact price history arrays for one or more market UUIDs, keyed by market ID. Pass a comma-separated list of UUIDs in the ids query parameter. Sparkline data is used for inline price trend charts in market listings.

​
Data freshness

All market data comes directly from the ingestion pipeline. Predexy ingests from all seven platforms on a rolling schedule, so the data you receive is always fresh from the latest scan cycle. There is no stale-data warning on the markets endpoints — if a platform is temporarily unavailable during ingestion, that platform’s data reflects the last successful fetch.
The market_count field on each DiscoverItem from the discover endpoint tells you at a glance how many platform markets are linked to that canonical question. Questions with market_count >= 2 have cross-platform pricing and are the most interesting for arbitrage research. See GET /api/v1/discover for full field documentation.