Schemas Get a free key
LIVE --:--:-- UTC Hyperliquid positions snapshot on a 5-min UTC grid — last boundary --:-- next in -:--
Prediction Markets

Polymarket Markets API

Live queryable Polymarket data. The /v1/polymarket REST namespace serves market discovery and detail, one token's tape summary, candles, order-book snapshots, the raw event tape and on-chain settlements, plus per-wallet positions and PnL and smart-money aggregation — the last two also as MCP tools.

Coverage: Markets, tape, order books, candles, settlements, wallet profiles and smart-money signals. Free — API key — delivered via MCP · API.

Data summary

CategoryPrediction Markets
CoverageMarkets, tape, order books, candles, settlements, wallet profiles and smart-money signals
DeliveryAPIArchiveStreamMCP
AccessFree — API key
MCP toolspolymarket_markets polymarket_market_detail polymarket_ohlcv polymarket_wallet polymarket_smart_money
REST/v1/polymarket/*

Getting started

bash
# Substring match on the market slug; active=true keeps open markets only.
curl -H "X-API-KEY: vk_..." \
  "https://data.vike.io/v1/polymarket/markets?query=bitcoin&active=true"

Example applications

  • Smart-money screening

    If you believe historically profitable wallets are early on mispriced markets, polymarket_smart_money surfaces where they are concentrating before you commit research time to a market.

  • Wallet due diligence

    If a wallet keeps showing up on the right side of resolutions, polymarket_wallet gives you its positions and PnL history — enough to separate skill from one lucky election.

  • Market discovery for archive studies

    If you are about to replay a market from the L2 archive, /v1/polymarket/markets and its detail route resolve slugs, condition_ids and token ids — the join keys the Parquet files are organized by — and /settlements gives the on-chain resolution.

Access

Free — get an API key at vike.io/account (up to 3 keys per account) and send it as X-API-KEY on every request.

Archive files are licensed CC BY 4.0 — use them for anything, credit Vike. API responses carry the same attribution ask. Keys, rate limits and errors: the API page.

FAQ

How does this relate to the archive datasets?
The API answers "what is happening / what happened, summarized" without downloading anything; its event tape and order books start 2026-07-25, like the archive, which holds the same recording as daily Parquet files for tick-level replay. They share identifiers (condition_id, token_id), so moving between them is a join.
What are the rate limits?
One free key carries one hourly request budget, shared by its REST and MCP calls; the number, and how keys and OAuth apps count, are on the API page (/data/api).