📖

L2 Order Book Snapshots

20-level L2 orderbook snapshots from HyperLiquid at ~100ms resolution

About This Dataset

Full L2 orderbook snapshots with 20 levels of bids and asks from HyperLiquid. Captures depth profile, spread, and imbalance at approximately 100ms intervals.

Partitions

exchange: hyperliquid

Schema

Column	Type	Description
`time_chain`	`timestamp`	Exchange snapshot timestamp (UTC)
`time_local`	`timestamp`	Ingestion timestamp (UTC)
`coin`	`string`	Asset symbol
`bid_prices`	`list[float64]`	Array of 20 bid prices
`bid_sizes`	`list[float64]`	Array of 20 bid sizes
`ask_prices`	`list[float64]`	Array of 20 ask prices
`ask_sizes`	`list[float64]`	Array of 20 ask sizes
`spread_bps`	`float64`	Bid-ask spread in basis points

Use Cases

Reconstruct the orderbook at any point in time
Compute orderbook imbalance and depth metrics
Detect liquidity voids and support/resistance zones
Model market impact for large order execution

R2 Path

s3://algotick-data-lake/states/deep_book/exchange=hyperliquid/year=YYYY/month=MM/day=DD/node={node}/data.parquet

Replace YYYY, MM, DD with the target date. Data is collected from node=eu-central.

Query with DuckDB

import duckdb

df = duckdb.sql("""
    SELECT *
    FROM read_parquet(
        's3://algotick-data-lake/states/deep_book/exchange=hyperliquid/year=2026/month=04/day=20/node=eu-central/data.parquet'
    )
    LIMIT 100
""").df()

print(f"Rows: {len(df)}, Columns: {list(df.columns)}")

Download via API

import requests

resp = requests.get(
    "https://algotick.dev/v2/history/raw?dataset=book&date=2026-04-20",
    stream=True,
)

with open("orderbook.parquet", "wb") as f:
    for chunk in resp.iter_content(8192):
        f.write(chunk)

# Then query locally
import duckdb
df = duckdb.sql("SELECT * FROM 'orderbook.parquet' LIMIT 100").df()
print(df)

Get API Key →

Related signals: Whale Activity Liquidations

Don't just stare at the dashboard. Automate it.

Every metric on this page is available via our sub-millisecond API.
Build trading bots, backtest strategies, and power AI agents with institutional-grade data.

Explore API →

Continue the Research

⚡ API Reference 🍳 Cookbook 📊 Backtests 📖 Playbooks