REST API

Flow and microstructure

What the order book and the order flow say about the balance of power. Five families: cumulative volume delta, aggregated depth, weighted average prices, directional pressure and liquidation zones.

#CVD

GET/v1/cvddata_type · cvd_<tf>

GET/v1/cvd/historydata_type · cvd_<tf>_history

The cumulative volume delta: the running sum of market buy volume minus market sell volume. It measures who, of the buyers or the sellers, had to cross the order book, hence who is really pushing the price.

Parameter	Type	Default	Description
`symbol`	string	required	The asset.
`timeframe`	string	required	1m, 5m, 15m, 30m, 1h, 4h, 1d.
`live`	boolean	false	Exposes the window in progress.
`since_ms / until_ms`	integer	—	On the history only.
`limit`	integer	100	On the history, from 1 to 1000.

curl -H "X-API-KEY: $BYTNODE_KEY" \
  "https://api.bytnode.com/v1/cvd?symbol=BTCUSDT&timeframe=1h"

Field	Type	Description
`bucket`	string	Start of the window.
`cvd_series`	float[]	The cumulative series inside the window, one point per internal step, from oldest to newest. It is its slope that carries the information.
`net_delta`	float	Buy minus sell over the whole window.
`taker_buy_volume / taker_sell_volume`	float	The two volumes that make up the delta.
`points_count`	integer	Number of points in the series.
`is_closed`	boolean	False only with live=1.

#Order book

GET/v1/raw/orderbookdata_type · raw_orderbook

The last known state, aggregated: the latest snapshot of each venue, then the sum of the depths. No limit: the route returns a single object.

Field	Type	Description
`timestamp`	string	The most recent of the states retained.
`bid_total`	float	Bid volume, all venues.
`ask_total`	float	Ask volume, all venues.
`imbalance`	float \| null	(bid − ask) / (bid + ask), computed on the aggregated totals and not averaged per venue. Positive: buyers dominate.

GET/v1/orderbook/aggregateddata_type · orderbook_aggregated_<tf>

GET/v1/orderbook/aggregated/historydata_type · orderbook_aggregated_<tf>_history

The same depth, but summarised by window: mean, extremes and dispersion. This is the useful form for a chart or a regime detection, since an instantaneous order-book state is too noisy for that.

Field	Type	Description
`bucket`	string	Start of the window.
`bid_total_avg / _min / _max`	float	Cumulative bid volume, on average and at the extremes.
`ask_total_avg / _min / _max`	float	The same on the ask side.
`imbalance_avg / _min / _max`	float	The imbalance over the window.
`imbalance_std`	float \| null	Its standard deviation: a measure of the stability of the balance of power.
`samples_count`	integer	Number of samples in the window.
`is_closed`	boolean	False only with live=1.

#VWAP

Two VWAPs, not to be confused. The first is cumulative since midnight UTC: it is the average price of the day. The second is that of a window: one value per candle. Both are weighted by volume and aggregated across venues.

curl

# The VWAP of the day, since midnight UTC
GET /v1/vwap?symbol=BTCUSDT

# The VWAP of the last closed hour, one value per window
GET /v1/vwap/window?symbol=BTCUSDT&timeframe=1h

GET/v1/vwapdata_type · vwap

GET/v1/vwap/historydata_type · vwap_history

Field	Type	Description
`vwap`	float \| null	Volume-weighted average price since midnight UTC.
`cumulative_volume`	float	Cumulative volume, which serves as the weight.
`timestamp`	string	Timestamp of the last computation.

This route does not carry is_closed: it is not a candle but a running value.

GET/v1/vwap/windowdata_type · vwap_window_<tf>

GET/v1/vwap/window/historydata_type · vwap_window_<tf>_history

Field	Type	Description
`bucket`	string	Start of the window.
`vwap`	float \| null	Volume-weighted average price over the window.
`total_volume`	float	Volume of the window.
`trade_count`	integer	Number of trades in the window.
`is_closed`	boolean	False only with live=1.

#Directional pressure

GET/v1/buysell/ratiodata_type · buysell_ratio

GET/v1/buysell/ratio/historydata_type · buysell_ratio_history

Field	Type	Description
`timestamp`	string	Timestamp of the computation.
`buy_volume`	float	Buy volume over the 5-minute window.
`sell_volume`	float	Sell volume.
`buy_ratio`	float	buy / (buy + sell), from 0 to 1. Above 0.5, buyers dominate.

GET/v1/trades/avg-sizedata_type · trade_size_avg

GET/v1/trades/avg-size/historydata_type · trade_size_avg_history

The average size of a trade profiles the participants: it rises when institutional players come in, it collapses when the flow becomes a dust of retail orders.

Field	Type	Description
`avg_size`	float \| null	Average size, weighted by the number of trades of each venue.
`total_volume`	float	Total aggregated volume.
`trade_count`	integer	Aggregated number of trades.
`timestamp`	string	Timestamp of the computation.

#Liquidation heatmap

GET/v1/heatmap/clustersdata_type · heatmap_clusters

GET/v1/heatmap/clusters/historydata_type · heatmap_clusters_history

The zones where potential liquidations concentrate, above and below the price. They are zones of attraction: the price goes there to find the liquidity they represent.

Field	Type	Description
`timestamp`	string	Timestamp of the computation.
`density_up / density_down`	float	Density of the most attractive cluster above and below the price.
`dist_up / dist_down`	float	Fractional distance of the cluster: 0.01 means 1%.
`attraction_up / attraction_down`	float	density / distance². The square makes the near weigh far more than the far.

How to read it

If attraction_down clearly exceeds attraction_up, the price is pulled downwards: liquidations of long positions are near. The reading is symmetrical in the other direction. It is an indicator of where the price has a mechanical reason to go, not a prediction.