Summary: Kalshi’s live order book is now available on DoubleZero Edge, a low-latency market-data transport layer. It is the first time Kalshi’s market data has been distributed over a multicast network, the first prediction-market venue on Edge, and the second venue overall after Solana. Starting today, every sports event contract and crypto perpetual futures market representing Kalshi’s most actively traded categories, with full book depth, is available to Edge subscribers on one low-latency connection.

Real Asset Classes Require Real Infrastructure

Real Asset Classes Require Real Infrastructure

The landscape of price discovery is shifting fast, and prediction markets have rapidly evolved into a primary financial asset class. Today, the prices on Kalshi represent live, tradable probabilities for everything from economic indicators and policy moves to sports events and crypto perpetual futures.

As serious volume continues to flow into these markets, individuals, quantitative desks, prop traders, and market makers all face the same problem: they need institutional-grade market data transport for signal generation. To answer that demand, engineers at DoubleZero and Malbec Labs have built the rails necessary for Kalshi’s order book data to ride on DoubleZero Edge. This effort represents a first for New finance, and is a major indicator of prediction market maturation as the need for high-performance connectivity continues to grow. Here’s what launched, why it matters, and what changes now.

Improving Signal Generation for Prediction Market Traders

So what: the first prediction market to deliver data like an exchange

Until now, the only way to get Kalshi data was the public API everyone else connects to. Kalshi’s REST API does its job well for reference data and snapshots, a signed WebSocket for live deltas and trades. It is the right tool for a trading app, a dashboard, or a research notebook. A latency-sensitive strategy run at scale is a different kind of job, and three properties of any public-internet API path matter for it:

It is accessed through a CDN. You land on the nearest CDN edge, so the distance from the engine to you sits outside your control.

It is unicast and metered. Every consumer is a separate connection served one copy at a time, inside a request budget. Scaling coverage or adding machines pushes against that model.

It leaves normalization to you. Kalshi quotes a Yes side and a No side. Turning that into one clean book per market, seeded from snapshots and kept in sync, is work every consumer repeats on its own.

Multicast delivery over dedicated, low-latency fiber has previously existed only behind institutional walls. With Edge, that changes. Edge delivers data via multicast, not unicast. Adding a subscriber, or adding markets to a group, costs the publisher almost nothing. It is the same delivery model proven on DoubleZero’s Solana shreds feed, where the network delivers data faster than the public internet and arrives first on more than 90% of slots for participating leaders.

Connecting to Kalshi via DoubleZero Edge unlocks three massive structural advantages for those that require faster and more consistent access to market data like serious individual traders, quantitative desks, market makers, and prop traders:

A lower latency floor. The feed is sourced directly from Kalshi and travels dedicated fiber instead of traversing a CDN edge over the public internet. For probability prices that move on news, the time you save is the edge.

Breadth without penalty. The liquid market set arrives on one feed with no per-stream cap and no per-request budget to manage. Watch every market that matters, on every machine that needs it, at a cost that does not climb with each new subscriber.

A clean, deterministic feed. One consistent Yes-side book per market, fixed-point prices, a self-describing reference channel, and built-in recovery. You spend your time on signal, not on reconciling quotes or rebuilding state. And because it is the same format as every other Edge venue, a cross-venue strategy integrates once and extends for free.

“Traditional finance got this concept exactly right: data access is a critical part of market structure… DoubleZero Edge brings institutional-grade infrastructure into the hands of prediction-market traders with software-defined access. Kalshi coming online shows that Edge optimizes market data for any venue, onchain or off.”

— Austin Federa, Co-founder of DoubleZero

How It Works

Delivered by multicast. The publisher emits once. The DoubleZero network replicates it to every subscriber simultaneously using multicast technology. No per-stream limits. No server-level fan-out.

Bypasses the public internet. A direct, low-latency path from Kalshi to Edge subscribers over dedicated fiber.

Yes/No normalization. Kalshi quotes both a Yes and a No side for every contract. The feed normalizes them into one consistent book per market. Subscribers see one clean view without reconciling two sides.

Always in sync. Snapshots cycle continuously in the feed. Join late or drop a packet – full book state rebuilds from the feed itself, no separate snapshot channel needed.

Who is This For

Built for traders who’d otherwise build it themselves.

Market Makers Skip building custom infrastructure to parse Kalshi’s JSON book and track new listings. Get a complete book, spend your time on the strategy.

Professional Traders Full L2 depth with no minimum volume levels, or lightweight L1 for tracking price action. Get the picture you need, so you can spend your API tokens on execution.

Bot & Arb Builders One connection instead of custom per-market subscriptions and manual book reconstruction.

The Future of Prediction Market Data

Now what: open infrastructure for new finance

Traditional finance settled this question decades ago: data access is a critical part of market structure. The world’s largest trading firms built dedicated networks to move market data fast and reliably but those networks are closed, centralized, and slow to join, and they were built only for themselves.

New finance grew up without that infrastructure. Prediction markets in particular have become a real source of financial alpha: the price of a Kalshi contract is a live, tradable probability for an event, and the quants, market makers, and prop desks that treat it that way want institutional-grade access to the data. Firms that wanted a sequenced, machine-readable view of a prediction market have generally had to build it themselves, assembling book state from API responses and running their own infrastructure to do it.

Read the full DoubleZero blog here 

Source: DoubleZero