Indexed HyperliquidBlockchain Data

Hyperliquid is a Layer 1 appchain that has been producing blocks since 2023, and it exists to run one application: the Hyperliquid exchange. Where a typical chain hosts a trading venue as a set of smart contracts, Hyperliquid implements the venue in the chain itself. Placing an order, cancelling it, matching it and paying funding are state transitions executed by the protocol, on a validator set running HyperBFT, a BFT consensus in the HotStuff family, with sub-second block times. That state machine is called HyperCore, and it is what this page covers. The native asset is HYPE. HyperCore has no EVM: contracts cannot be deployed to it. A paired EVM execution layer, HyperEVM, was added in February 2025 and shares the same consensus and the same block production, which makes Hyperliquid one of the few networks where two execution environments advance in lockstep. Spot markets run in the same state machine as perpetuals, under Hyperliquid’s own token standard rather than as deployed contracts, so listings and trades are protocol events too.

SonarX indexes Hyperliquid’s chain history and normalizes it into the same model used across every network SonarX covers, so Hyperliquid records can be joined to activity on other chains without a translation layer. Because the exchange is part of the protocol rather than a contract deployment, the interesting records are chain state rather than decoded logs, and they are indexed as such.

Full historical coverage is available today as a data-share or a file delivery. Real-time delivery keeps the same dataset current to chain tip. Hyperliquid has no token-balance surface to index: user state lives as positions inside the exchange state machine, so balance products do not apply here. Kafka streaming carries the same records as a live topic today.

Liquidations are the record teams ask about most, and on this chain they are marked on the fill that closes the position rather than emitted as a separate log. Hyperliquid liquidations data covers the marker and the datasets it joins to, and the liquidations, fills, funding and open interest and perp positions pages list the selected fields of each dataset.

Three things make this chain unlike an EVM L1. The exchange state has no event-log equivalent, so an indexer cannot subscribe to transfers and reconstruct the rest; it has to follow the protocol’s own state transitions. Blocks arrive faster than a second, which changes the economics of a streaming pipeline more than the total data volume does. And the network carries two execution environments on one consensus, so HyperCore and HyperEVM have to be indexed as separate datasets that nonetheless share a block height, or cross-surface analysis silently mismatches.

Hyperliquid requests come from research and risk teams more often than from application developers. Market-structure research on a fully onchain order book is the recurring one: funding and basis series, liquidation clustering, and maker and taker behavior studied across the venue rather than sampled through an API. Flow analysis is the second, following capital between HyperCore, HyperEVM and bridges to see where positions are actually funded from. Third is reconciliation, where a desk with exposure on the venue needs its own records to agree with the chain’s. None of that work needs an API: it is range queries over a long history, which is what a warehouse share is for.

Teams comparing venues usually pull Injective alongside it, and DeFi analytics work spans both surfaces of the network.

Tell us which surface you need, HyperCore or HyperEVM or both, how far back, and where the data should land. History is available now as a share or a file delivery, and streaming is provisioned on request. The published coverage matrix for every network SonarX indexes lives in the data documentation, and the rest of the catalogue is at all indexed chains.