All benchmarks
AlternativesSourced from Prediction market API rate limits, tested with a daily ramp

Polymarket alternatives

Prediction market platform with a CLOB API and public WebSocket

Polymarket runs a prediction market platform with a CLOB API (`/book`, `/midpoint`) and a Gamma markets endpoint, fronted by Cloudflare and accompanied by a public market WebSocket. Documented rate limits sit on the order of 1500 requests per 10 seconds for the book endpoint, but the Cloudflare layer queues bursts rather than rejecting them, so a polling loop or trading bot only sees rising latency before it ever sees a 429. Builders sizing against this kind of API need two numbers nobody publishes: the real latency of the hot endpoints at a polite request rate, and what actually happens as the request rate climbs. This benchmark probes book, price and list endpoints of five prediction market venue APIs continuously from three Railway regions (us-east, eu-west, sgp), every 5 to 7 seconds over warm keep-alive connections, with a separate cold-connect probe once per minute that forces a fresh TCP and TLS handshake. Once per day per venue it runs a 60-second-per-tier ramp at rising request rates, recording added latency versus the same-hour warm baseline and any 429 or 5xx onset. Polymarket's ramp tiers are 25/50/100 requests per 10 seconds (at most 7 percent of the documented book budget) with an automatic abort the moment throttled plus 5xx responses exceed 1 percent of a 10-second window. CDN cache hits are flagged from `cf-cache-status`, `x-cache` and `age` headers and excluded from latency aggregates so no venue ranks on its CDN. The cohort competing with Polymarket includes Kalshi (documented token bucket), Limitless (undocumented limits, errors CDN cached for 4 hours), Manifold (500 req/min per IP) and Myriad (30 req/10s keyless budget, single-region Heroku origin).

Positioning

Polymarket runs a hybrid Central Limit Order Book where orders match off-chain and settle atomically on Polygon through an audited exchange contract, with orders expressed as EIP-712 signed messages rather than exchange IOUs. Authentication is layered: L1 EIP-712 signatures derive API credentials from the user's private key, then L2 HMAC-SHA256 headers (POLY_ADDRESS, POLY_SIGNATURE, POLY_TIMESTAMP, POLY_API_KEY, POLY_PASSPHRASE) sign subsequent trading requests. Its architectural moat is non-custodial CLOB throughput on Polygon: the operator cannot execute unauthorised trades or set prices, and the matching engine sits primarily in eu-west-2 with co-location available after KYC. For a pm-rate-limits benchmark, the surface to size against is the CLOB /book and /midpoint endpoints plus the Gamma markets endpoint, all fronted by Cloudflare. The Cloudflare layer queues bursts instead of returning 429, so a polling loop sees rising latency before it sees an error, which changes what a throttle test can actually measure. Prediction-market builders, permissionless market creators and retail flow outside US jurisdiction integrate on Polymarket natively. Teams migrate to a regulated venue when they need CFTC-supervised counterparty, FIX connectivity or venue-side compliance guarantees the on-chain settlement model does not provide.

About Polymarket: polymarket.com

Polymarket across benchmarks

Top 5 Polymarket alternatives

  1. PredictIt logo

    PredictIt

    #1 · Warm book latency

    102msp99 1.12 s
  2. Kalshi logo

    Kalshi

    #2 · Warm book latency

    121msp99 397 ms
  3. Limitless logo

    Limitless

    #3 · Warm book latency

    285msp99 906 ms
  4. Manifold logo

    Manifold

    #4 · Warm book latency

    325msp99 594 ms
  5. Myriad logo

    Myriad

    #5 · Warm book latency

    518msp99 2.08 s
Prediction market API rate limits, tested with a daily ramp benchmark ledger. Warm book latency per provider, live measured data.
ProductWarm book p50Trend
Color24h
01
Polymarket logo
91 ms
02
PredictIt logo
102 ms
03
Kalshi logo
121 ms
04
Limitless logo
285 ms
05
Manifold logo
325 ms
06
Myriad logo
518 ms
Best
91 ms
Median
285 ms
Worst
518 ms
Spread
8.4×248 ms → 2.08 s
Samples · 24h
261,6578 providers

Warm book latency · last 24 hours

Warm book latency · last 24 hours

Show
openchainbench.com0200ms400ms600ms−24h−18h−12h−6hnowPolymarket96 msPredictIt104 msKalshi125 msManifold251 msLimitless297 ms

Interactive bar chart, distribution view, per-chain drilldowns and chart export on the benchmark page.

Open benchmark

Measurements sourced from Prediction market API rate limits, tested with a daily ramp. Methodology and raw data on the benchmark page.