The last traded price describes someone else’s fill. A larger order can reach several levels. This tool walks current Hyperliquid asks for a buy or bids for a sell, then computes a quantity-weighted average.
Read the result correctly
Choose base-coin quantity or USD notional; neither input is margin. The markets here are native Hyperliquid perpetuals, not spot order books. A notional input walks price levels until that amount is covered; for a sell it represents target gross proceeds, not a guaranteed payout. Price impact is measured against the best bid/ask midpoint, so it includes the spread and the cost of walking deeper levels. It is different from a maximum slippage tolerance entered in a trading app.
If the requested order exceeds visible depth, the average applies only to the covered portion. The remaining quantity has no estimated fill price. Snapshots older than two minutes require a refresh; even a fresh book can change before a real order reaches the exchange.
Check the arithmetic with two levels
Suppose the first ask offers one unit at 101 and the second offers two at 103. Buying two costs 204 and averages 102. Buying four exceeds the three visible units; there is no defensible price for the fourth from this snapshot alone.
Those numbers are illustrative. Live data above has a retrieval time, and the CSV contains the exchange’s snapshot timestamp.
Download the same snapshot
The CSV includes venue, exact market, UTC time, side, price, base-coin quantity and aggregate order count. Sum quantities down each side to reproduce the result. One snapshot is not a complete historical Level 2 feed.
Download the field dictionary · Download the read-only collector
python3 orderbook-snapshot.py --coin BTC > btc-book.csvThe collector uses only the public information endpoint. It needs no wallet, private key or trading permission. Running it later naturally produces a different snapshot.
What remains outside this estimate?
Trading fees, funding and transfer costs are separate. Once you use the estimated execution price in the break-even calculator, set that calculator’s extra slippage assumption to zero to avoid counting the same impact twice.