Hyperliquid Data for Market Makers: Book History, L4, Flow

Making markets on Hyperliquid is different from making markets on a centralized exchange. The book is on-chain, every order belongs to a visible address, and every trade can be attributed. That transparency is an edge, but only if you have the data to exploit it: the history of the book to backtest your quoting, the identity of the flow that hits you, and the funding and liquidation context around it.

Here is what Hypedexer provides for market makers and quantitative trading teams.

1. Order book history, since April 2024

Backtesting a quoting strategy needs the book as it was, not just the trades.

GET /market/book/{coin}?at=<timestamp> returns the order book at any point in time:

• April 2024 to February 2026: the top ~20 levels per side, reconstructed minute by minute (source=minute), plus full snapshots about every 10 minutes (source=snapshot).
• Since September 2026: up to 50 levels per side, rebuilt from the full order book of our own Hyperliquid node.

curl -H "X-API-Key: $HYPEDEXER_KEY" \
  "https://api.hypedexer.com/market/book/BTC?at=2024-08-05T06:30:00Z&levels=20"

That is enough to measure spreads and depth through past volatility events (the August 2024 crash, token listings, liquidation cascades), calibrate your quote sizes against real queue depth, and replay how the book refilled after a sweep.

2. The live order-level book (L4), with owners

Most venues give you an aggregated book: a price and a total size per level. Hyperliquid's node knows every individual order and who placed it. Hypedexer streams that order-level book over WebSocket:

{"method": "subscribe", "subscription": {"type": "l4Book", "coin": "BTC"}}

You receive a snapshot of every resting order (order id, owner address, price, size), then every change to the book. With it you can:

• Know your queue position at each level, not estimate it.
• See which market makers are quoting, how much size each one shows and how often they requote.
• Detect spoofing and layering: orders that appear and vanish before they can be hit.
• Spot large resting orders from known wallets before they trade.

See l4Book and l4BookUpdates in the docs.

3. Who is hitting you: flow attribution

Every Hyperliquid trade has a buyer and a seller address. Hypedexer serves fills linked to wallets from 1 January 2024, almost 247 million trades in 2024 alone and more than 4 billion in total.

For a market maker, that turns your fill log into a map of your counterparties:

• Toxicity by wallet: measure the mark-out of trades against each taker address, and widen or skew against the ones that consistently trade ahead of moves.
• Flow segmentation: separate retail flow, other market makers, vaults and liquidations.
• Competitor analysis: follow other liquidity providers' quotes in the L4 book and their volume through their recent fills.

Query any address with GET /fills/user/{address}, or the whole market with GET /fills/ filtered by coin and time. Details in our wallet history guide.

4. Funding, liquidations and execution context

• Funding rates per market since 12 May 2023 (GET /funding/fundingHistory): hourly rate and premium, to model carry and the cost of holding inventory.
• Liquidations (GET /liquidations/): size, price, liquidated address and the liquidation method, to anticipate cascades.
• Priority gas per fill (since April 2026): who pays to jump the queue, and how much.
• Builder fees and TWAP parent ids on fills served through POST /info.

5. Your own reporting

Market makers need clean records too. Hypedexer publishes daily account statements for any address: balances, positions, realized and unrealized PnL, fees, funding, transfers, one row per day, as JSON or CSV. Useful for fund administrators, auditors and proof-of-inventory reports. See Daily account statements.

Access

• REST and WebSocket on one key, in EU and Asia regions: create one in the console.
• Historical data before March 2025 is available on the Scale and Enterprise plans. WebSocket connections and subscriptions scale with the plan: see pricing.
• For bulk delivery, custom datasets or a dedicated setup, talk to us.

FAQ

Is there historical order book data for Hyperliquid? Yes. Hypedexer serves the book at any point in time from April 2024 (top ~20 levels per side), rebuilt from the full order book of our own node since September 2026. The order-level (L4) book, with each order's owner, is streamed live over WebSocket.

What is the L4 order book on Hyperliquid? The list of every individual resting order with its id and owner address, as opposed to L2, which aggregates size per price level. Hyperliquid's node maintains it; Hypedexer streams it over WebSocket.

Can I see who traded against me on Hyperliquid? Yes. Every trade has a public buyer and seller address. Match your fills against the market tape by time and trade id to identify each counterparty.

Do you have Hyperliquid funding rate history? Yes, hourly, per market, since 12 May 2023.