Data catalog
Three layers built from raw Hyperliquid fills: the raw tiles — order-flow OHLCV, funding and positioning — the forecast layer — a funding/volatility forecast and wallet-behaviour (toxicity) factor set you stack into your own
model — and market intelligence — whole-market wallet-flow regime signals. Most datasets are partitioned per (coin, month); the market-wide ones carry one row per
month. Each is served as a Parquet file — open any dataset to see what every column means.