Pyth Network

At a glance

What it does

A first-party publisher is a firm that submits prices it sees in its own trading, rather than a node that scrapes an exchange's public API. Pyth is built on that idea. Publishers send prices to an oracle program on Pythnet, Pyth's own application chain, which combines them into one aggregate price plus a confidence interval, a band that widens when publishers disagree.

The pull model is how those prices reach other chains. On every Pythnet slot, validators post the Merkle root of all prices; Wormhole guardians sign that root. An off-chain service called Hermes serves the latest signed prices to anyone. When an application needs a price, the user's transaction submits the update, and the Pyth contract on that chain checks the Wormhole signatures and the Merkle proof before storing it. Nothing is written until someone asks, so a quiet asset costs nothing to keep listed. The documented fee is one unit of the chain's smallest denomination, such as 1 wei on Ethereum, meant to pay data providers.

Pyth Pro, previously Pyth Lazer, is a subscription product with API keys for low-latency users. Pyth Indices, launched in June 2026 with Marketvector, offers 24/7 indices for US equities, metals and oil.

RWA-relevant work

Pyth prices the market side of real-world assets: stocks, ETFs, FX and commodities such as gold and silver. In September 2025 it partnered with Blue Ocean Technologies, an SEC-registered alternative trading system, to publish overnight US equity prices on-chain. In July 2024 Ondo and Pyth launched a USDY/USD feed across more than 65 chains. On 28 August 2025 the US Department of Commerce chose Pyth, alongside Chainlink, to publish official GDP data on-chain.

Pyth says it is "the only oracle that offers real-time access to US equities"; that is a company claim. A community forum post claiming Pyth prices about 95% of on-chain RWA perpetuals volume is unofficial and undated.

How it is different

Chainlink and RedStone run their own node networks that fetch data. Pyth pays the data owners to publish directly, and gives applications a confidence interval along with the price, which a lending protocol can use to be more careful when markets are split.

The limit shows in fund NAVs. A money-market fund or credit fund has no trading desk quoting it every second; its value comes from an administrator once a day. In the sourced material Pyth publishes no NAV feed for a tokenized fund. That work sits with RedStone, Chainlink's NAVLink and Chronicle. Pyth's place in RWA is the traded asset: the tokenized stock, the gold token's reference price, the FX rate.

Where it fits

Getting prices on-chain compares push and pull models and explains why market prices and NAVs need different oracles. Tokenized stocks and Robinhood Chain shows why equity feeds after market hours matter. Who's who in RWA places Pyth in the data layer. Related profiles: Ondo, Robinhood.

Risks and open questions

Pull oracles move a duty onto the application. If nobody submits an update, the stored price stays old, and each protocol must enforce its own freshness limit. A careless integration can read a stale value without noticing.

Publisher quality varies. The aggregate is only as strong as the set of firms quoting a given asset, and for thin assets that set can be small. Pyth's confidence interval makes the disagreement visible but does not remove it.

Pyth holds no financial licence for its data products; the Data Association is a Swiss non-profit association. That is normal for oracles but matters as regulated funds start to depend on its prices. No incident specific to Pyth's RWA or government-data feeds was found in public reporting.

What to watch next

• Use of overnight and weekend equity prices by tokenized-stock venues, which tests whether 24/7 trading of a stock that closes at 4 p.m. can be priced fairly.
• Growth of Pyth Pro subscriptions, the clearest sign of paying institutional demand.
• Whether Pyth adds NAV or proof-of-reserve products, which would put it in direct competition with RedStone and Chainlink for tokenized funds.
• More public-sector data after the Commerce Department deal.