Prediction Markets Odds API – Polymarket, Kalshi, Manifold
Pricing
from $2.00 / 1,000 market snapshots
Prediction Markets Odds API – Polymarket, Kalshi, Manifold
Reliable odds feed: never a silent empty result. A source that is down gives a clear free error row; no match gives a free no_results row. Live Polymarket, Kalshi and Manifold odds in one schema: probability, bid/ask, 24h volume, liquidity, close time.
Pricing
from $2.00 / 1,000 market snapshots
Rating
0.0
(0)
Developer
Liora Blyth
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Share
Prediction Market Odds API – Polymarket, Kalshi & Manifold in One Schema
Live odds from the three biggest prediction markets – Polymarket, Kalshi and Manifold – normalised into one schema: implied probability (yes price), bid/ask, 24h volume, liquidity, close time and URL. Search by keyword ("fed", "bitcoin", "election") or pull the top markets by volume.
- ✅ Never a silent empty dataset: if a source is down you get a free
status: "error"row naming it; if nothing matched you get a freestatus: "no_results"-type row saying so. Errors and status rows are never charged. - ✅ Official open-data APIs only – no scraping of websites, no login, no proxies needed
- ✅ One clean, documented schema across all sources
- ✅ Pay only for results: $0.002 per market ($2.00 per 1,000). Errors are free.
Coverage
- Sources: Polymarket (gamma-api.polymarket.com), Kalshi (api.elections.kalshi.com trade API v2), Manifold (api.manifold.markets v0) – public read-only APIs, no key.
- Markets: open markets only; worldwide topics (politics, economics, crypto, tech, science, weather…). Kalshi sports are skipped unless you pass
kalshiCategories/keywords. - Fields: platform, marketId/slug, event, question, outcome label + all outcomes, yes/no price (implied probability), best bid/ask, last price, 24h price change, 24h and total volume, liquidity, open interest, currency, real-money flag, category, status, close/created time, resolution source, URL, fetchedAt. Fields a platform does not publish are
null. - Not covered: order books, trade history, resolved-market archives, Betfair/PredictIt.
When a source is down – error, not empty
- Each platform is fetched independently, with retries on HTTP 429/5xx (honouring
Retry-After). - If a platform still fails, you get one free row
{"platform": "Kalshi", "status": "error", "error": "…"}and the other platforms continue. The run still ends SUCCEEDED, so check forstatusrows. - If the platforms answered but nothing matched your keywords/volume filter, you get one free row
{"status": "no_results", "error": "Polymarket, Kalshi, Manifold answered, but no open market matched …"}– never a silent empty dataset. - Market rows never have a
statusoferror/no_results, sostatus in ("error","no_results")separates the two.
Cost example and cost cap
- 3 platforms × 100 markets (default) = up to 300 markets = $0.60 + $0.00005 start. 50 keyword matches ≈ $0.10.
- Cap it two ways:
maxItems(total markets, e.g.100→ max $0.20) and/or Max charge per run (Console → Run options, ormaxTotalChargeUsdin the API). When the charge limit is hit the Actor stops cleanly and keeps what it saved. Status rows are free.
What can you use it for?
- Compare the same event across venues and spot price gaps (arbitrage / research)
- Feed dashboards, newsletters and trading bots with probabilities on macro, crypto, politics and tech
- Track market sentiment over time with Apify schedules (every 5 min, hourly, daily)
- Academic / journalism research on forecasting accuracy
Input
| Field | What it does |
|---|---|
keywords | e.g. fed, bitcoin, trump – empty = top markets by 24h volume |
platforms | polymarket, kalshi, manifold (default all) |
maxMarketsPerPlatform | Billable results cap per venue |
maxItems | Optional total billable cap for the whole run (empty = no total cap) |
minVolume24h | Skip illiquid markets |
kalshiCategories | e.g. Economics, Crypto, Politics (default: all non-sports) |
kalshiMaxEvents | How many Kalshi events to expand |
Example input:
{"keywords": ["fed","bitcoin"],"platforms": ["polymarket","kalshi","manifold"],"maxMarketsPerPlatform": 25}
Output
Each result is one row in the dataset (download as JSON, CSV, Excel or via API). Example:
{"platform": "Polymarket","marketId": "2589813","marketSlug": "will-the-fed-increase-interest-rates-by-25-bps-after-the-october-2026-meeting-20260617190324032","eventId": "606422","eventTitle": "Fed Decision in October?","question": "Will the Fed increase interest rates by 25 bps after the October 2026 meeting?","outcomeLabel": "25 bps increase","outcomes": [{"name": "Yes","price": 0.045},{"name": "No","price": 0.955}],"yesPrice": 0.045,"noPrice": 0.955,"bestBid": 0.04,"bestAsk": 0.05,"lastPrice": 0.04,"priceChange24h": -0.03,"volume24h": 784045.448559,"volumeTotal": 14393749.342364,"liquidity": 1546726.8007,"openInterest": null,"currency": "USDC","isRealMoney": true,"category": null,"status": "open","closeTime": "2026-10-29T03:59:00Z","createdTime": "2026-06-17T23:05:57.870064Z","resolutionSource": "https://www.federalreserve.gov/monetarypolicy/fomccalendars.htm","url": "https://polymarket.com/event/fed-decision-in-october-20260617190323537","fetchedAt": "2026-10-24T16:41:52+00:00"}
Resolution rules next to the odds
Markets with the same question can resolve differently on each platform. Every row includes rulesText (Polymarket description, Kalshi primary + secondary rules, Manifold description; whitespace collapsed, max 2,000 characters) and, for Kalshi, canCloseEarly. Read the rules before treating two markets as the same bet. No platform login is needed; only public endpoints are used.
Pricing
Pay-per-event, no subscription:
| Event | Price |
|---|---|
Market snapshot (market) | $0.002 ($2.00 / 1,000) |
| Actor start | $0.00005 |
Error rows and empty results are never charged. Set Max charge per run in Apify Console to cap your spend – the Actor stops cleanly when the limit is reached.
Option names (shared across our open-data Actors)
The five Actors in this pack (Prediction Markets, US Building Permits, EU Company Registry, UK Public Tenders, EU Fuel Prices) use the same names for the same job:
| Job | Option | Notes |
|---|---|---|
| Free-text filter | keywords (array) | EU Company Registry uses query (one company name) + registrationNumbers |
| Date window | daysBack + …From / …To | issuedFrom/To (permits), publishedFrom/To (tenders) |
| Total billable cap | maxItems | available in all five; per-source caps (maxMarketsPerPlatform, maxItemsPerCity, maxResultsPerCountry, maxPagesPerSource) still work |
| Only new rows on schedules | onlyNewSinceLastRun | EU Fuel Prices: onlyChangedSinceLastRun (prices that changed) |
| Which sources | platforms / cities / countries / sources | the natural unit of each source |
Status rows always use the same status values: error / source_unavailable (source down), no_results / no_match / not_found (source answered, nothing matched), no_new (nothing new since last run). They are never charged.
FAQ
Is the data official?
Yes – it comes from each venue's public, read-only market-data API (no login, no wallet, no trading). The Actor never places orders.
Are prices probabilities?
yesPrice is the last/mid price of the YES contract on a 0–1 scale, which is the market-implied probability. Manifold uses play money (isRealMoney: false).
Why is Kalshi slower?
Kalshi lists ~16,000 open events; the Actor filters them first and then fetches markets per event (5 in parallel) to avoid the tens of thousands of combo markets.
Do you collect user data?
No. Creator / trader names are dropped; only market-level data is returned.
Can I get historical prices?
Not yet – schedule the Actor to build your own time series. A price-history mode is on the roadmap.
Legal & data sources
This Actor only reads public data published by government bodies or by the platforms' own public APIs under open licences (each row has source/license or url fields where available). It returns organisation-level data and deliberately excludes personal data. You are responsible for using the data in line with the source licence and applicable law (e.g. GDPR).
Integrations
Schedule runs, then push results to Google Sheets, Slack, webhooks, Zapier, Make or your database with Apify integrations, or call the Actor from the Apify API / Python / JavaScript clients.