Hyperliquid has quickly become one of the leading decentralized perpetual futures exchanges, combining a custom blockchain with trading speeds that match centralized platforms. What makes it unique is its dual architecture: Hypercore handles the order book and trading, while HyperEVM lets developers build smart contracts on top of that liquidity. In this guide, we'll explain how Hyperliquid works, what data it provides, and how Dwellir's infrastructure can help you access it all without running your own nodes.

Quick answer: Hyperliquid is a custom blockchain built specifically for perpetual futures trading. It uses HyperBFT consensus to achieve fast transaction speeds, Hypercore to run its order book, and HyperEVM to support Ethereum-compatible smart contracts. Dwellir's Hyperliquid services provide you with reliable access to all of Hyperliquid's data: real-time WebSocket feeds, RPC endpoints, historical data, and dedicated node infrastructure.

What Makes Hyperliquid Different

• Custom blockchain: Hyperliquid runs on its own layer-one blockchain built around HyperBFT consensus, which is based on HotStuff and optimized for fast finality and reliable order matching.
• Fully onchain order book: Unlike most decentralized exchanges, Hypercore keeps every perpetual and spot order directly on the blockchain, maintaining price-time priority just like traditional exchanges.
• Smart contract support: HyperEVM provides an Ethereum-compatible environment using the same validators, so developers can build applications that tap into Hypercore's trading liquidity.
• High performance: The network currently processes around 200,000 orders per second with room to scale even higher.

Why This Matters for Developers

1. True transparency: Because the matching engine runs within the blockchain consensus, there are no hidden off-chain order queues or relayers - everything is visible and verifiable.
2. Shared state: Hypercore and HyperEVM work together, which means smart contracts can read live trading data and execute trades in the same transaction - perfect for building vaults, automated strategies, and monitoring tools.
3. Rich data access: You get multiple data layers (order book depth, trade feeds, historical archives) that provide detailed market insights without needing to scrape websites or reverse-engineer APIs.

How Hyperliquid Works: The Three Core Components

HyperBFT: The Consensus Layer

HyperBFT is Hyperliquid's consensus mechanism, based on the HotStuff algorithm. Here's what you need to know:

• Byzantine fault tolerant: The network can handle up to 1/3 of validators being malicious or offline while continuing to operate normally.
• Fast finality: Transactions achieve finality in about 0.2 seconds on average (under 0.9 seconds even at the 99th percentile), which is fast enough to support professional trading strategies.
• Built to scale: The consensus layer is already designed to handle millions of orders per second as the execution layer continues to optimize.

Hypercore: The Trading Engine

Hypercore runs the actual order books for every market on Hyperliquid. It works like a traditional exchange's matching engine, but entirely on-chain:

• Unified markets: Both spot and perpetual markets use the same settlement system, which simplifies risk management and margin tracking.
• Fair ordering: Orders match using price-time priority (first come, first served at each price level) with all the tick size and lot size rules you'd expect from a professional exchange.
• Predictable execution: Each block processes actions in a specific order (non-order actions, then cancellations, then new orders) to minimize reordering issues.

HyperEVM: Smart Contract Platform

HyperEVM adds Ethereum compatibility to Hyperliquid, letting you deploy Solidity smart contracts that can interact directly with the trading data:

• Two block types: Fast "small" blocks every ~2 seconds handle routine transactions, while "big" blocks arrive every minute with 30M gas for more complex operations.
• Shared accounts: Your wallet works across both trading and smart contracts - no bridges needed between Hypercore and HyperEVM.
• Fee structure: Uses EIP-1559 style fees (base fee + priority fee) with automatic burning to keep costs predictable.

Comparing Hypercore, HyperEVM, and the Order Book Server

Understanding Hyperliquid's Data Layers

Real-time streams

• Trades: Every completed trade with details like buy/sell side, size, and participant information is available through both public WebSocket APIs and Dwellir's hosted server.
• L2 order book depth: Aggregated bid and ask prices at different levels, useful for monitoring spreads and building trading strategies.
• L4 order book depth (private): Individual order-level data showing who placed each order, only available through the order book server for the most detailed market view.

Batch & historical datasets

• S3 archives: Hyperliquid publishes periodic snapshots to S3 buckets, providing raw historical data for backtesting and analysis.
• Normalized datasets: We clean and format the data from Hyperliquid's evolving schemas through our historical service, making it easier to query and analyze. Check out the managed trade datasets.
• gRPC snapshots: The GetOrderBookSnapshot method provides secure, point-in-time views of the order book, useful for compliance and trading analysis.

Rate limits vs. dedicated access

The public Hyperliquid infrastructure limits you to 100 WebSocket connections, 1000 subscriptions, and 1200 REST requests per minute. While this works fine for testing and small projects, it becomes restrictive when you're running multiple trading bots or monitoring many markets simultaneously. Dwellir's dedicated endpoints remove these limits, providing much higher capacity with geographic redundancy.

Why Use Dwellir for Hyperliquid Data

1. Lower latency: Our hosted order book server reduces median latency by 51 ms compared to public feeds, which is a 24.1% improvement based on benchmarks across 2,662 trades.
2. Access to private L4 data: The co-located server provides detailed order-level data (L4) that shows individual user orders, which isn't available through public endpoints.
3. Everything in one place: Access HyperEVM RPC, gRPC, WebSockets, and historical data through a single account with transparent pricing.
4. Reliable infrastructure: We handle connection management, data compression, and monitoring so your feeds stay up even when nodes restart.

Dwellir's Hyperliquid Service Map

1. Low-Latency WebSocket Access

Our hosted order book server is located right next to a Hyperliquid validator, which minimizes the distance your data has to travel. Trade updates, order book changes (L2), and detailed order data (L4) reach you faster through a private connection. The service includes automatic compression and connection management to keep your feeds running smoothly.

What's included:

• Authenticated access to trades, L2 order book, and L4 detailed order data (docs)
• Monitor many markets through a single connection
• Automatic connection health checks to prevent disconnections
• Private L4 data showing individual user orders for detailed analysis

2. HTTP and RPC Endpoints

We provide access to Hyperliquid's HTTP APIs and HyperEVM RPC endpoints, so you can execute trades, check balances, or interact with smart contracts without running your own node. HyperEVM access is available through our API Platform, and full Hypercore RPC is available via dedicated Hyperliquid nodes.

Common uses:

• Track your positions, funding rates, and margin status in real-time
• Build trading bots that route orders across multiple venues
• Trigger automated actions like liquidations or rebalances based on market data

3. gRPC Streaming and Snapshots

While WebSockets are great for continuous updates, gRPC provides reliable sequencing and point-in-time snapshots. We offer authenticated methods like GetOrderBookSnapshot and streaming block fills, which help you save state checkpoints, verify trade execution, or recover after connection issues.

4. Historical Data and S3 Archives

Dwellir offers managed historical datasets for Hyperliquid order book snapshots, trade logs, and transaction fills, with extended retention and ready-to-ingest formats for warehouses like Snowflake, BigQuery, or ClickHouse.

Common use cases:

• Regular imports for backtesting strategies or compliance records
• Building features for machine learning models (order book imbalance, user trading patterns, liquidation risk)
• Post-trade analysis to verify execution quality

5. Dedicated Nodes and Execution Environments

For strategies that need the absolute lowest latency, we can set up dedicated Hyperliquid validator and archive nodes along with private VPS instances in the same data center. This lets you run your trading algorithms and risk systems right next to the nodes serving your data, minimizing latency and ensuring consistent performance. Contact us to discuss your requirements.

What's included:

• Private Hypercore and HyperEVM nodes with SLA support
• Optional direct connections to your existing infrastructure
• Real-time monitoring, alerts, and 24/7 support

Getting Started with Dwellir

1. Set up your account: For HyperEVM RPC, sign up at our dashboard. For the order book server, gRPC streams, or historical data access, contact our team. Need Hypercore RPC? Explore our dedicated Hyperliquid nodes.
2. Connect via WebSocket: Start with the Python code examples from the Hyperliquid order book server repository, then build out your own clients with reconnection logic.
3. Import historical data: Sync historical data from S3 into your data warehouse, or use Dwellir's pre-processed datasets.
4. Add gRPC calls: Use GetOrderBookSnapshot and other gRPC methods for reliable snapshots and state management.
5. Build on HyperEVM: Deploy smart contracts and connect them to live market data using the same credentials.

Common Use Cases

• Market making: Use detailed order-level data (L4) to see who's placing orders, track spreads in real-time, and analyze historical fill patterns to improve your quoting strategy.
• Quantitative research: Combine historical snapshots from S3 with gRPC snapshots to reconstruct order book states and train trading models.
• Risk monitoring: Build dashboards that pull data from both HyperEVM smart contracts and the order book to track leverage, positions, and funding rates.
• Compliance and auditing: Archive trade fills and order book states by block number to document execution quality and maintain audit trails.

Sample WebSocket Subscription (Python)

import asyncio
import json
import os
import websockets

SUBSCRIPTION = {
    "method": "subscribe",
    "subscription": {
        "type": "l2Book",
        "coin": "ETH",
        "nLevels": 20,
        "nSigFigs": 5,
    },
}

async def stream_l2():
    url = os.environ["DWELLIR_HYPERLIQUID_WS"]
    async with websockets.connect(url, extra_headers={"x-api-key": os.environ["DWELLIR_API_KEY"]}) as ws:
        await ws.send(json.dumps(SUBSCRIPTION))
        async for message in ws:
            data = json.loads(message)
            top_bid = data["data"]["bids"][0]
            top_ask = data["data"]["asks"][0]
            print(f"ETH bid {top_bid['px']} / ask {top_ask['px']}")

if __name__ == "__main__":
    asyncio.run(stream_l2())

Tip: For the lowest latency, run this client on Dwellir's dedicated execution environment, which keeps median response times under 20 milliseconds.

FAQ

Is Hyperliquid just another EVM chain? No, it's more specialized. Hyperliquid combines Hypercore's professional-grade order matching engine with HyperEVM's Ethereum-compatible smart contracts, all secured by the same consensus layer. This means trading and smart contracts work together seamlessly.

Can I access L4 order book data without Dwellir? Not through public endpoints. L4 data (detailed order-level information) is only available through Hyperliquid's order book server. Dwellir's hosted version adds lower latency, 5x order book depth, and we handle load balancing, upgrades, and monitoring so you don't have to.

How do public rate limits affect me? Hyperliquid's public infrastructure limits you to 100 WebSocket connections and 1200 REST requests per minute per IP. For production trading systems that need more capacity, Dwellir's dedicated endpoints remove these limits.

Where should I start? For HyperEVM RPC, sign up at our dashboard. For the order book server, contact our team. Then try the open-source Python examples and gradually add gRPC snapshots and smart contracts as your system grows.

---

Ready to start building on Hyperliquid? For HyperEVM RPC access, sign up at our dashboard. For the order book server with L4 data, contact our team. See the Hyperliquid documentation for guides and API references.