How to get Kalshi historical data

Kalshi's API gives you past trades, one-minute candles and settled markets, free and without an API key. It does not give you past order books: the order book endpoint shows only the book as it is now.

Checked against Kalshi's API on October 31, 2026.

What Kalshi's API has

Every path is under https://external-api.kalshi.com/trade-api/v2. A live endpoint stops returning a record once it moves to the historical ones, so a full history reads both.

Every trade in a market

import time
import requests

API = "https://external-api.kalshi.com/trade-api/v2"

def pages(path, key, **params):
    """Every row of a paged endpoint, following the cursor to the end."""
    params["limit"] = 1000
    while True:
        r = requests.get(f"{API}{path}", params=params, timeout=30)
        if r.status_code == 429:  # rate limited: wait, then ask again
            time.sleep(1)
            continue
        r.raise_for_status()
        body = r.json()
        yield from body[key]
        params["cursor"] = body.get("cursor")
        if not params["cursor"]:
            break

ticker = "KXBTC15M-26OCT311945-45"
trades = [
    *pages("/historical/trades", "trades", ticker=ticker),  # before the cutoff
    *pages("/markets/trades", "trades", ticker=ticker),     # after it
]

One-minute candles for the same market, from 7:30 to 7:45 PM Eastern:

candles = requests.get(
    f"{API}/series/KXBTC15M/markets/{ticker}/candlesticks",
    params={"start_ts": 1793489400, "end_ts": 1793490300, "period_interval": 1},
    timeout=30,
).json()["candlesticks"]

for c in candles:
    print(c["end_period_ts"], c["yes_bid"]["close_dollars"], c["yes_ask"]["close_dollars"])

What the API leaves out

• Past order books. The depth, spread and size at each price at any past moment are gone once the book changes.
• Top of book between minutes. Candles give the YES bid and ask once a minute, not every change.

Order book history

Kalshi Eye records Kalshi's market feed around the clock: every order book change on every market, every trade and every settlement, 28 billion rows so far. Each UTC day is published the next morning as Parquet files.

• book_l2: every book change, to rebuild the whole book at any moment.
• book_snap_1s: the top ten levels each second.
• bbo and bbo_1m: every top-of-book change, and minute bars.
• trades and settlements.

from kalshieye import Client, replay_book  # pip install kalshieye

kx = Client()  # reads KALSHIEYE_API_KEY
book = kx.load("book_l2", "2026-10-13", "2026-10-13", tickers=["KXBTC15M-26OCT131200-00"])
for ts, ticker, bids, asks in replay_book(book):
    ...  # the whole book after every change

Questions

Next

• Kalshi historical order book data: datasets, plans and coverage.
• Data dictionary: every column of every dataset.
• How to backtest Kalshi markets
• How to get Polymarket historical data