FinFeedAPI Prediction Markets API
Nine event venues normalized into one REST schema, billed by usage credit.
by FinFeedAPI
Last updated
What it is
A normalized REST feed over nine event venues at once, sold by the request rather than by the seat. It is the broadest coverage in this category: Kalshi, ForecastEx, Crypto.com and Gemini on the regulated side, Polymarket, Myriad and Pascal on the decentralized side, Manifold as a play-money venue, and Hyperliquid's on-chain outcome markets.
The shape is a market-data API rather than a prediction-market one, which is the point — exchanges and markets are discovered through endpoints, then trades, quotes, current and historical order books and OHLCV candles hang off a market id, with the same field names whichever venue produced them. OHLCV periods run from one second to five years. Historical trades, quotes and raw limit order events come from flat files, addressed by a full-day query or a bounded time range, which is what makes an order-book replay possible at all.
The publisher is API BRICKS LTD, a UK company formerly called CoinAPI LTD; FinFeedAPI is the brand this product is sold under, and CoinAPI is its sibling. Usage credits are shared across both.
Availability
An account and an API key, no identity check and no geographic gate in the customer agreement, which is governed by the laws of England and Wales and was last effective 18 August 2025. Two clauses are worth reading before you build: access is granted for internal business purposes only, and the customer is solely responsible for any licence the underlying venue requires for the data it supplies. A venue's own terms can restrict what you do with its prices regardless of what you paid FinFeedAPI.
An unauthenticated request to the prediction-markets host answers 401 and names the four header forms it will accept, so the service is live and gated rather than gone.
Pricing
Usage credits, one credit being one US dollar at the pay-as-you-go rate. Requests are billed at 5 USD per 1,000 for the first 1,000 requests in a day and 1 USD per 1,000 above that, so the marginal cost falls sharply once a client is polling in volume and a light, occasional integration pays the worst rate.
Committed plans at 64, 256, 512 and 1,024 USD a month buy credits at 0.85, 0.75, 0.70 and 0.65 USD respectively. The vendor states plainly that a commitment does not reduce usage — it buys more credits for the same spend, and anything past them bills at the pay-as-you-go rate. All plans get the same endpoints; what escalates is support, and only the 1,024 tier carries a 99.9% uptime SLA.
Every plan sees the same data, there is no recurring free allowance, and the 25 USD of starter credits for a qualified new organization is a one-off. Credits do not expire; they are also not refundable. An extra API key above the plan's quota costs 1 USD a day, enabled or not.
Markets & resolution
Whatever the nine venues list, which in practice spans elections, sports, macro releases and
crypto prices. FinFeedAPI resolves nothing and settles nothing: it republishes each venue's market
records, so resolution_source on this card is null and every resolution question belongs to the
venue that wrote the rules.
One consequence of the breadth is worth stating. A single settlement value cannot describe a set
that contains a CFTC-regulated exchange, an on-chain market in USDC and a play-money site — the
card says cash for the regulated majority, and a normalized price from Manifold is not a price
anybody can be paid at. The venue is in every record; use it before treating two rows as
comparable.
Market ids are per outcome, not per question: a binary market has a distinct id for each tradable side. Anyone counting markets or joining across venues needs that in mind from the first query.
Integrations
Three interfaces over the same data: REST, a JSON-RPC proxy for the same read endpoints, and a
hosted MCP server for agents, which takes the key in an X-APIKey header and exposes tools for
exchange and market discovery, activity, OHLCV and order books. Responses come back as JSON or
msgpack.
Client libraries are generated from the OpenAPI description and published in the vendor's api-bricks-sdk repository under an MIT licence, covering Python, JavaScript, TypeScript, Go, Java, C# and a dozen more; the repository was last committed to on 18 September 2026. The SDK being open source does not make the service open source — the data is paid and hosted.
No WebSocket on this product. Live state is a snapshot you fetch.
Limitations
The billing model is the risk. A poller that touches the API once a second is 86,400 requests a day and is priced accordingly, so the cost of a naive integration is a function of your loop, not of your data volume. Model it before you commit, using the vendor's own worked pricing examples.
No streaming means no tick-level latency story, and a card that wants the fastest possible view of a Polymarket book will be better served going to the venue directly.
Coverage is a list of venues, not a promise about any one of them — depth, history and field completeness vary by exchange, and the vendor documents no per-venue backfill start date. Check the history you actually need on the venue you actually care about before the credits run down.
Alternatives
Adjacent covers only Kalshi and Polymarket but adds indices, cross-venue reference rates and news, and sells seats rather than requests. The venues' own APIs are free, deeper and streaming, at the cost of one integration per venue and a normalisation layer you maintain. Dome, the other unified API in this space, was acquired by Polymarket in February 2026 and its documentation sets end of life for the Dome APIs at 28 April 2026.
Specs
- Interfaces
- API, Python, MCP server
- Export
- JSON
- Available in
- Global
- KYC required
- No
- Market subjects
- Politics, Sports, Macro, Crypto
- Resolved by
- Resolves nothing
- Maker fee
- None
- Platforms
- Web, Library
- AI features
- None
- Capabilities
- Order book, Cross venue
- Pricing verified
- Availability verified
Background
How this part of the sector works, rather than which product to pick.
- Where weather data comes from and what you may do with it — Observations, model output and archives are three products with three licences. The forecast is the cheap half; a clean observation history is not.
- Why the same contract costs two different prices on two venues — Two venues quoting one outcome are quoting two contracts. What the gap between the prices is made of - rules, fallbacks, fees, collateral, exits, eligibility.
- What a trade actually costs on a prediction market — Venues publish trading fees in four incompatible units, so 3.00% on one can be cheaper than 1.75% on another. How to convert them, and what else takes a cut.
Also worth comparing
- Adjacent — Prediction-market indices and reference rates, with Kalshi on the tier you can test.
- Apify prediction-market scrapers — A store of third-party actors that normalize Kalshi and Polymarket into one row shape.
- Predictefy — Sixteen venues behind one verb family, priced in credits, with nothing readable for free.
- Kalshi API — REST, WebSocket and FIX access to a CFTC-regulated event exchange.
- Polymarket US API — A keyless public gateway for reading, and a signed key behind KYC for trading.
- DepthFeed — The recorded bid-ask ladder for crypto up-or-down markets, which no venue keeps itself.
Named as a replacement for
FAQ
Which venues does the FinFeedAPI Prediction Markets API cover?
Nine, grouped by the vendor as regulated (Kalshi, ForecastEx, Crypto.com, Gemini), decentralized (Polymarket, Myriad, Pascal), social (Manifold) and on-chain outcome markets (Hyperliquid). The exchange list is itself an endpoint, so the current set can be read rather than guessed.
Is there a free tier?
No recurring one. A qualified new organization gets 25 USD in usage credits once, and after that every request is billed — 5 USD per 1,000 for the first 1,000 requests a day, then 1 USD per 1,000. Credits do not expire and are not refundable.
Does it stream prices?
Not on this product. The three documented interfaces are REST, JSON-RPC and a hosted MCP server; live state is fetched as a snapshot. FinFeedAPI ships WebSocket feeds on other product lines, not on prediction markets.
Can I get historical order books?
Yes, as raw limit-order events (adds, updates, deletes) pulled from flat files for replay, alongside current order-book snapshots, historical trades and quotes, and OHLCV from one second to five years.