Updates
·
The Market Data Layer for New Financial Markets
Pyth ended Q3 at $11.49M active ARR, up 86.5% QoQ. Where the growth is going: exchanges, prediction markets, and AI.
Q3 was the quarter when Pyth’s commercial model began to show its shape.
Revenue, product usage, customer acquisition, and market coverage all moved in the same direction at the same time. Pyth is no longer used only to bring prices onchain. It is becoming infrastructure for new financial products across exchanges, prediction markets, and AI.
Q3 by the numbers
A commercial model beginning to scale
Q3 was not only a quarter of headline growth. It provided early evidence that Pyth is building a repeatable commercial model.
Total active ARR reached $11.49M at the end of September, up 86.5% quarter over quarter, after Pyth added $5.33M in net new ARR during Q3.
The business is also developing across multiple products. At the end of September, Pyth Pro represented $9.68M in live ARR. Pyth Indices represented $1.81M across 16 live clients. Across the business, Pyth has now 276 paying accounts with the terminal self-serve becoming another important route into the model with $500K in new ARR for September.
Where the growth is going
Pyth’s Q3 traction is broadening across three high-potential verticals: exchanges, prediction markets, and AI. In each one, current usage is an early proof point for a much larger opportunity.
Exchanges
Pyth is already part of the infrastructure behind a growing market for new exchange products.
During Q3, real-world-asset perpetual markets generated $2.09 trillion in volume, with Pyth-powered venues accounting for 94.1% of that activity. Coinbase is using Pyth data to support continuously priced thematic markets, while MarketVector indexes calculated with Pyth data underlie the equity index futures listed by Coinbase Derivatives.
Other leading exchanges powering their markets with Pyth data (Pro and Indices) include Trade[XYZ], OKX, and Lighter.
Exchanges are leveraging Pyth to launch products that would be difficult to support with traditional market-data infrastructure. As these markets expand across more assets, venues, and geographies, the opportunity for Pyth expands with them.
Prediction Markets
Prediction markets need clear, continuous data inputs to price markets and settle outcomes. Pyth is already supplying that infrastructure to leading platforms.
Kalshi is the headline example. The platform uses Pyth prices for markets covering stocks and commodities, and in September named Pyth as the sole Source Agency in filings for gold and silver perpetuals on its CFTC-registered exchange.
Pyth also powers Polymarket’s real-world-asset markets.
As prediction markets move beyond elections and news events into financial assets, commodities, and macroeconomic outcomes, reliable market data becomes central to the product itself. Pyth is positioned to provide that data layer as the category keeps growing.
AI and Agentic Finance
AI systems will need structured, current, and historical financial data that can be retrieved and used programmatically.
Pyth and Exa are testing what that market-data layer for financial agents could look like. Exa provides web-scale research context. Pyth provides structured current prices, historical observations, and candlestick data through MCP. Together, they allow an agent to research a company, asset, or event, retrieve the relevant market data, and combine both into a single answer.
The opportunity extends well beyond answering market questions. As agents begin to monitor assets, evaluate opportunities, manage risk, and interact with financial products, real-time market data will become a core input into software-driven finance. Pyth can provide that input across a wide range of assets and use cases.
Across all three verticals, the pattern is the same: Pyth is already embedded in products that are being built today. As more exchanges, prediction markets, and AI systems move from experimentation into production, Pyth’s usage grows with the markets it enables.
Nasdaq Basic on the Data Marketplace
Pyth is now an approved external distributor of Nasdaq Basic, Nasdaq’s proprietary real-time quote and trade product for US equities.
Nasdaq Basic will be available through the Pyth Data Marketplace, which is built for software and blockchain-native destinations. Clients interested in Nasdaq Basic must license the product directly from Nasdaq and receive prior written approval before consuming the feed.
Read the full announcement here.
The 100% Rule
The PYTH DAO has approved the 100% Rule.
Every dollar the DAO receives from Pyth products will now go toward buying PYTH on the open market, up from one-third of its non-PYTH balance under the previous framework.
That represents a threefold increase in the share committed, at the same time as the revenue behind it is growing.
The full detail, including how acquisitions are executed and reported, is covered in Pyth Goes All In.
From here
Pyth has moved from being known primarily for crypto and onchain applications to becoming a cross-asset market data layer for exchanges, prediction markets, institutions, applications, and AI systems.
The next phase is to convert more Pyth Pro trials into production usage, expand institutional adoption, launch more indices, grow supply and demand on the Marketplace, deepen integrations with exchanges and prediction markets, and turn AI interest into commercial deployments.
The early signals are visible across the business: ARR growth, deeper account expansion, new exchange products, growing prediction-market adoption, and the first applications of Pyth data in agentic finance.
Q4 is about turning that early traction into deeper adoption, more products, and more market activity.
Missed the Pyth Connect Q3 Livestream? Watch it in full here.
Q3 was the quarter Pyth's commercial model began to show its shape. Active ARR reached $11.49M, up 86.5% quarter over quarter, across 276 paying accounts. This recap covers where that growth is going (exchanges, prediction markets, and AI), Nasdaq Basic on the Data Marketplace, and the DAO's approval of the 100% Rule.
