Prediction Markets API - Polymarket, Kalshi & Manifold Odds

Get live prediction markets from Polymarket, Kalshi and Manifold in one dataset: the question, current probabilities for every outcome, 24-hour and total volume, liquidity, close date, a unified category and the link to the market. Filter by keywords, category, volume, liquidity or closing date; the same question listed on several platforms is cross-linked so you can compare odds side by side.

The Actor reads only the platforms' public read-only APIs — no API keys, no wallets, no scraping, no proxies. It never places trades.

What you can do with it

• Odds dashboards and newsletters — "what does the market think about the Fed, the election, Bitcoin, this weekend's games" refreshed hourly.
• Cross-platform comparison — the alsoOn field lists the same question on other platforms with its probability there; spot gaps between Polymarket, Kalshi and Manifold.
• Event research — pull every market closing this week in a category, or every market mentioning a company, country or person.
• Data for AI agents — expose forecasts to an agent through the Apify MCP server ("what is the probability of X according to prediction markets?").
• Monitoring — schedule runs and push changes to Slack, Sheets or a database with Apify integrations.

Platforms

Platform	Currency	Coverage	Notes
Polymarket	USD	Events with nested markets, tags, 24h volume, order-book best bid/ask	Sorted by 24h volume
Kalshi	USD	Regulated US exchange: politics, economics, weather, sports, culture	Category from Kalshi itself; combo/provisional markets skipped
Manifold	MANA (play money)	Community markets on anything, thousands of questions	Volume and liquidity are in MANA, not dollars

Input

Field	Type	Default	Description
keywords	array	[]	Whole-word match in question, event title, tags and outcome names. Empty = everything.
excludeKeywords	array	[]	Drop markets containing any of these.
category	select	all	politics, crypto, sports, weather, economics, science-tech, culture, other.
platforms	array	all three	Any of polymarket, kalshi, manifold.
minVolume / minLiquidity	number	0	Skip thin markets.
closesWithinDays	number	—	Only markets resolving within N days.
sortBy	select	volume24h	volume24h, volume, liquidity, closesAt, probability.
maxItems	number	500	Cap after filtering; platforms are interleaved fairly at the cap.
includeClosed	boolean	false	Include closed/resolved markets.
includeDescription	boolean	false	Add resolution rules / description text.

Example — hottest crypto markets closing this month, at least $10k volume:

{ "category": "crypto", "closesWithinDays": 30, "minVolume": 10000, "sortBy": "volume24h", "maxItems": 200 }

Output

{
  "id": "polymarket:4434785",
  "platform": "polymarket",
  "platformName": "Polymarket",
  "title": "Will the Fed cut rates in September?",
  "event": "Fed decision in September?",
  "url": "https://polymarket.com/event/fed-decision-in-september",
  "marketType": "binary",
  "outcomes": [{ "name": "Yes", "probability": 0.81 }, { "name": "No", "probability": 0.19 }],
  "probability": 0.81,
  "lastPrice": 0.81,
  "bestBid": 0.8,
  "bestAsk": 0.82,
  "volume": 2203656.01,
  "volume24h": 50231.4,
  "liquidity": 33603.98,
  "openInterest": null,
  "currency": "USD",
  "category": "economics",
  "platformCategory": null,
  "tags": ["fomc", "fed-rates", "economic-policy"],
  "status": "open",
  "closesAt": "2026-09-17T00:00:00Z",
  "createdAt": "2026-08-01T12:00:00Z",
  "imageUrl": "https://…",
  "alsoOn": [{ "platform": "manifold", "url": "https://manifold.markets/…", "probability": 0.7 }],
  "scrapedAt": "2026-09-10T15:02:11Z"
}

probability is the Yes price for binary markets and the leading outcome for multi-outcome markets. Kalshi prices are mid-market when there was no last trade. The OUTPUT record in the key-value store holds a run summary with per-platform counts and any platform errors (a run still succeeds if at least one platform answers).

Use cases

Ready-made examples you can run without configuring anything:

• Track Manifold Markets questions — ready-made task, run it as is
• Compare Polymarket and Kalshi odds — ready-made task, run it as is

Track Manifold Markets questions

Manifold's play-money markets with probabilities, volume and close dates in the same schema as the real-money venues.

{ "platforms": ["manifold"], "keywords": ["AI"] }

Compare Polymarket and Kalshi odds

The same question listed on both venues is cross-linked, so the spread between them is one subtraction away.

{ "platforms": ["polymarket", "kalshi"], "minVolume": 10000 }

Monitor event contracts odds

Filter by category, volume, liquidity or close date and run it on a schedule to build a time series of implied probabilities.

{ "platforms": ["polymarket", "kalshi", "manifold"], "category": "politics" }

Pricing

Pay per event: a small fee per Actor start plus a fee per market returned. Filtering is free — you pay only for markets in the dataset. Current prices are in the Pricing tab.

Good to know

• Categories come from the platform where it provides one (Kalshi categories, Polymarket tags); otherwise a keyword heuristic assigns them. Use keywords when you need precision.
• Manifold uses play money; treat its volume as engagement, not dollars.
• Polymarket sports markets (thousands of match-winner markets) dominate 24h volume. Use category or excludeKeywords to hide them.
• Data is a snapshot at run time. For history, schedule the Actor and keep each run's dataset.

Use with AI agents (MCP)

Add the Actor to the Apify MCP server and ask an agent: "What do prediction markets say about a rate cut this month?" Typical tool input: { "keywords": ["rate cut", "fed"], "maxItems": 20 }.

Related Actors

• SEC EDGAR API
• Hacker News Search API
• Bulk URL Status Checker

Disclaimer

Not affiliated with Polymarket, Kalshi or Manifold. Market data belongs to the platforms; this Actor only reads their public APIs. Nothing here is financial advice.

Cost control

Pay per event only works if you can see the bill before you pay it:

• Set dryRun to count every result and print what a full run would cost - nothing is charged, and only a few sample rows are stored.
• You are charged only for rows actually delivered to the dataset; a run that finds nothing costs nothing beyond the start event.
• maxItems is respected exactly, and the run stops before the platform's Max total charge for the run.
• Every run writes a COST record to the key-value store: chargeable events, charged events, the amount in USD and whether a limit was hit.