Market data

#Market data

#Candles

Returns OHLCV candle data for one market. Results are paginated by time. Weight: 2.

{
  "type": "candles",
  "symbol": "PERP_BTC_USDC",
  "interval": "1h",
  "start_time": 1716100000000,
  "end_time": 1716200000000,
  "limit": 500
}

Each item in data.rows contains open, high, low, close, and volume as decimal strings, plus timestamp as the candle-open time in milliseconds. volume is base-asset volume. data.next_cursor is null on the final page. Responses are cached for 60 seconds using the symbol, interval, time range, and cursor as the cache key.

#Funding comparison

Returns current and rolling-average funding-rate figures by market and venue. Weight: 2. This response is not paginated and is cached per symbol for 60 seconds (up to 256 entries).

{
  "type": "fundingComparison",
  "symbol": "PERP_BTC_USDC"
}

data.rows contains a market symbol, next_funding_time (millisecond epoch), and exchanges[]. Each venue entry contains name, last, and the 1d, 7d, and 30d average rates. Rates are decimal strings. The venue name is an identifier, such as binance, bybit, or pangoperps.

#Funding rate history

Returns historical funding rates and the mark price captured at each funding interval. Records are written at the 00:00, 08:00, and 16:00 UTC boundaries. Weight: 2; cache duration: 30 seconds.

{
  "type": "fundingRateHistory",
  "symbol": "PERP_BTC_USDC",
  "start_time": 1716100000000,
  "limit": 500
}

Each row includes funding_rate (decimal string for the 8-hour interval), mark_price (decimal string at the funding boundary), and funding_rate_timestamp (UTC millisecond epoch). The response includes data.next_cursor for pagination.

#Liquidations

Returns recent liquidation events across the platform. Each row includes position information and, when account lookup succeeds, the affected address and account. Weight: 2.

{
  "type": "liquidations",
  "symbol": "PERP_ETH_USDC",
  "limit": 50
}

Each row includes symbol, side (LONG or SHORT), position_qty, notional, mark_price, est_liq_price, address, account_id, broker_id, and timestamp. Decimal quantities/prices are strings; timestamp is milliseconds. The address/account/broker fields can be null if lookup fails. Account enrichment is cached for 60 seconds. data.next_cursor is null when no further results are available.

#Market detail

Returns one combined snapshot for a market: market information, orderbook, recent trades, funding history, and candles. The subqueries run in parallel. Weight: 2.

{
  "type": "marketDetail",
  "symbol": "PERP_BTC_USDC",
  "include": ["market_info", "orderbook", "recent_trades", "funding_history", "candles"],
  "orderbook_levels": 20,
  "recent_trades_limit": 20,
  "funding_history_limit": 8,
  "candles_interval": "1h",
  "candles_limit": 24
}

data contains symbol and the selected sections. market_info uses the same fields as the marketSummary market row; orderbook uses the same book structure as orderbook; the other sections follow the corresponding marketTrades, fundingRateHistory, and candles row formats. A section excluded through include is absent from the response rather than set to null. Combined sections may have different freshness, and overall latency is determined by the slowest section.

#Market summary

Returns mark/index prices, 24-hour activity, funding, open interest, and trading metadata across markets. Weight: 1.

{
  "type": "marketSummary",
  "symbols": ["PERP_BTC_USDC", "PERP_ETH_USDC"]
}

data includes total_24h_volume, total_open_interest, and markets[]. Each market row can contain symbol, mark_price, index_price, 24h_open, 24h_close, 24h_high, 24h_low, 24h_volume, 24h_amount, open_interest, last_funding_rate, est_funding_rate, next_funding_time, bid_price, ask_price, max_leverage, min_notional, quote_tick, and base_tick.

Prices, volumes, rates, open interest, and symbol metadata are decimal strings. next_funding_time is a millisecond epoch. est_funding_rate may be null before the first estimate is received; aggregate volume/open-interest values may be null when unavailable. Market rows are sorted lexically for stable rendering.

Omitting symbols uses an in-memory snapshot of all active markets. Supplying symbols fetches those markets concurrently with a 10-second overall budget; if some lookups time out, successful market rows are still returned. There is no pagination.

#Market trades

Returns recent taker-side fills for one market. Each fill is enriched with the taker address, account, and broker when available. Weight: 1; no pagination.

{
  "type": "marketTrades",
  "symbol": "PERP_BTC_USDC",
  "limit": 50
}

Each row contains symbol, side (BUY or SELL, from the taker's perspective), executed_price, executed_quantity, address, account_id, broker_id, and executed_timestamp. Prices and quantities are decimal strings; the timestamp is milliseconds. The wallet address is returned in full. The query filters for taker fills so each trade appears once; per-row address lookups are cached for 60 seconds.

#Orderbook

Returns a market's current orderbook snapshot, with mid-price and spread. Weight: 1; no pagination.

{
  "type": "orderbook",
  "symbol": "PERP_BTC_USDC",
  "max_level": 100
}

data contains symbol, mid_price, spread, asks[], bids[], and an inner ts timestamp. Each level has decimal-string price and quantity. Asks and bids are ordered best-first and trimmed to max_level. The mid-price is (best ask + best bid) / 2 rounded to eight decimals using HALF_DOWN; spread is best ask - best bid. Both are null if either side has no levels. An empty book returns empty arrays, not an error. The snapshot is served from memory and is typically available in under a millisecond.