Recipe 52 of 65
How is a Hyperliquid coin's open interest split by position size and by account size?
/v1/hl-cohorts/symbols/{symbol}/size-breakdown splits one coin's open positions into ten position-size buckets; /cohort-metrics?axis=size splits the same book by the size of each wallet's account.
GET /v1/hl-cohorts/symbols/{symbol}/size-breakdown answers the position-size split in one call: ten rows for one
coin, one per position-size bucket from
>$2.5m down to <$1k: how many positions, their notional, the long notional, the notional close to liquidation and
the bias. It is the current book. For the split by how big each wallet's account is, use
GET /v1/hl-cohorts/symbols/{symbol}/cohort-metrics with axis=size. The full parameter lists are on
Coin size breakdown and
Coin cohort metrics.
Quick answer
Put the coin in the path; there are no other parameters to set.
# BTC open positions in ten position-size buckets, largest first
curl -H "X-API-Key: $VIKE_API_KEY" \
"https://data.vike.io/v1/hl-cohorts/symbols/BTC/size-breakdown"Example response
Captured from the live API on 2026-10-24 at 17:16 UTC, trimmed to the 4 largest of 10 buckets.
USD throughout. bias is (long - short) / (long + short) of the bucket's notional. On this capture 179 positions over
$2.5M held $2.45B, more than all other buckets together, and leaned short (-0.13), while every smaller bucket leaned
long, the <$1k bucket's 24,698 positions at +0.38.
[
{
"tier": ">$2.5m",
"min_size": 2500000.0,
"max_size": null,
"position_count": 179,
"total_position_value": 2454301613.0,
"total_position_value_long": 1068230711.0,
"value_close_to_liquidation": 68805380.39,
"bias": -0.1295033131
},
{
"tier": "$1m-$2.5m",
"min_size": 1000000.0,
"max_size": 2500000.0,
"position_count": 197,
"total_position_value": 314140435.5,
"total_position_value_long": 193145750.1,
"value_close_to_liquidation": 23904019.33,
"bias": 0.2296777384
},
{
"tier": "$500k-$1m",
"min_size": 500000.0,
"max_size": 1000000.0,
"position_count": 286,
"total_position_value": 203099496.5,
"total_position_value_long": 125010069.0,
"value_close_to_liquidation": 15557663.28,
"bias": 0.231022934
},
{
"tier": "$250k-$500k",
"min_size": 250000.0,
"max_size": 500000.0,
"position_count": 449,
"total_position_value": 160226373.4,
"total_position_value_long": 109945114.8,
"value_close_to_liquidation": 14239093.51,
"bias": 0.3723722569
}
]Variations
By account size instead. axis=size groups the same positions by the wallet's account-value tier (shrimp up to 4x whale). hours=1
returns the newest settled hour, one row per tier.
# BTC positioning per account-size tier, newest settled hour
curl -H "X-API-Key: $VIKE_API_KEY" \
"https://data.vike.io/v1/hl-cohorts/symbols/BTC/cohort-metrics?axis=size&hours=1"On the same capture the 16:00 UTC hour had 22 positions held by 4x whale accounts worth $1.16B with bias -0.58,
against 29,075 shrimp positions worth $58M at +0.34.
The position-size split as an hourly series. axis=tier is the same ten buckets as size-breakdown, hour by hour.
# one week of BTC positioning per position-size bucket
curl -H "X-API-Key: $VIKE_API_KEY" \
"https://data.vike.io/v1/hl-cohorts/symbols/BTC/cohort-metrics?axis=tier&hours=168"The bucket bounds. The ladders and their dollar bounds, so the labels need no guessing.
# every cohort ladder with its bounds
curl -H "X-API-Key: $VIKE_API_KEY" \
"https://data.vike.io/v1/hl-cohorts/segments"Before you rely on it
Position size and account size are different ladders
size-breakdown and axis=tier bucket each position by its own notional. axis=size buckets by the wallet's account
value, so a small position of a large account sits in a whale tier. The tier names differ too: dollar ranges for
positions, shrimp to 4x whale for accounts. Read /v1/hl-cohorts/segments for both ladders' bounds.
Close to liquidation means past halfway
value_close_to_liquidation is the notional of positions whose mark has moved more than half of the way from entry to
liquidation price. Empty buckets come back as rows of zeros, so the table always has ten rows, largest first.
Ask your agent
Connected to the Vike MCP server (https://vike.io/mcp) in Claude, ChatGPT or Cursor, ask it in plain words.
The hl_market tool answers with view: size_breakdown; hl_cohort_metrics with axis: size gives the account-size split, at 1 credit a call.
How is BTC open interest on Hyperliquid split by position size?