Crypto market data, with research standards.
OHLCV across 1,500+ instruments — every series undergoes the highest data-quality validations and is checked continuously by automated data-quality agents. The first commercially-available ML feature store for crypto markets. Options analytics most retail platforms charge five figures a year for — volatility surfaces, higher-order Greeks, IV-vs-RV spreads, pin candidates. A growing risk-indicator library — correlation grids, multi-confidence VaR, drawdown curves, regime tags. Cross-venue funding rates. On-chain data features no one else productizes — NVT, address velocity, whale activity, network congestion. Crypto-native market breadth. US macro joined directly to the crypto tape. No SDK, no enterprise sales call. Plain JSON, CSV, or XLSX.
why this exists
The crypto data industry built itself around two extremes — toy free APIs that drop pairs without notice and rate-limit you at 60 calls a minute, and Bloomberg-priced terminals that exist to be expensed by a desk that already has one. The CDD API was built for the audience neither side serves: independent analysts, academic researchers, small funds, and disciplined retail investors who want institutional depth without the markup, and reliability without the multi-year contract.
The real difference is the surface area you couldn't buy at this price before. The first commercially-available ML feature store for crypto — pre-computed regime tags, momentum cohorts, volatility regimes, and microstructure features built directly off the underlying tape. A growing risk indicator library — correlation grids, multi-confidence VaR, drawdown surfaces, market regime labels. Institutional analytics, productized.
stale data is the enemy
Six layered checks every subscriber gets that a free CSV download does not — plus a quiet set of proprietary validators that catch what the disclosed checks miss. This is what we run, when we run it, and what happens when a row fails.
-
Bi-daily gap analysis
Every timeseries is scanned for missing intervals at 00:00 and 12:00 UTC. Gaps trigger an automatic re-pull from source. Un-fillable gaps — exchange outage, missing trading window, halted contract — are surfaced in the response payload so your backtest sees the discontinuity instead of silently inheriting it.
-
Stale-data detection
Every OHLCV row is reconciled against the exchange's own canonical record at ingest. Discrepancies above a per-field tolerance trigger a re-pull and an alert to the data team; rows that fail the second pull sit in quarantine until they're confirmed or corrected. Your subscription does not see provisional or out-of-date values.
-
Duplicate detection
Every row is hashed at ingest against its (symbol, interval, timestamp) key. Same-key duplicates are deduped before the warehouse write; partial-match anomalies — a row with the same timestamp but different OHLC values — trigger a divergence review. You get one canonical row per interval, every time, no matter how many times the upstream replays it.
-
Outlier flagging
Statistical filters watch every series — rolling z-scores on returns, IQR bands on volume, ratio tests on tradecount. A close that's 2× the prior bar, a volume print 100× the trailing median, or a tradecount that doesn't match its volume — all flagged for human review before ship. Genuine market events get confirmed and pass through; data errors get quarantined and re-pulled.
-
Cross-source consistency
For symbols traded on multiple venues, prices and volumes are reconciled against alternate sources within tolerance windows. Disagreements escalate to manual review. Real cross-exchange spreads — when they exist — are tracked as a separate feature in /data/features/ml/xexchange-spread/, not silently patched into a single "blended" series.
-
Proprietary validators
A library of internal checks we don't fully publish — pattern detection for prearranged trades, timestamp drift across the venue's own ledger, fee-percentile anomalies, miner-revenue versus hashrate divergence, and a few that catch things competitors don't even know to look for. The disclosed checks above are what we'll talk about; the proprietary set is part of why our data behaves better in production. Catalog grows with every post-mortem.
one tier · one flat price
No volume meters. No per-call billing. No enterprise sales call. The whole API surface at one flat price — markets, derivatives, funding, risk, on-chain, options analytics, and the ML feature store. Every live endpoint included; no held-back tier.
Every endpoint. Every format. Full feature store.
- Every live endpoint — no held-back tier
- Full ML feature store · 26 signals, growing
- 1,500+ instruments · all granularities
- JSON · CSV · XLSX output formats
- Personal commercial license
- Email support from the team
Cancel from your profile whenever; resume whenever. · Browse all endpoints →
six pillars · one subscription
The catalog organizes around six pillars — historical markets, derivatives, risk indicators, the ML feature store, on-chain data features, and funding & macro. Every endpoint shares the same authentication, the same parameter conventions, and the same three output formats. You learn the API once.
Spot, futures, and tick-level depth
OHLCV for 1,100+ Binance spot pairs, 205+ USDT-M futures, 125+ COIN-M futures, and 135+ Deribit assets. Daily, hourly, and 1-minute intervals on every series. Tick aggregations (610 / 1500 / 4500) for the majors. Some series go back to 2017; new listings folded in within a week.
Greeks, IV surfaces, funding
Full options chains for the four BTC/ETH/LTC/XRP majors with Δ Γ ν Θ, mark IV, bid/ask IV, and open interest in both contracts and USDT. Perpetual funding rates at 1H and 8H cadence. Deribit DVOL volatility indices daily. Top-100 option transactions per day ranked by USD notional — the tape institutions watch.
Correlations, VaR, drawdowns
A growing library of computed risk surfaces. Pairwise correlations and rolling correlations across every Binance pair (Pearson / Kendall / Spearman, 1w and 1m windows). Standalone 1-day VaR at 99 / 95 / 90% for 1,100+ symbols, plus by-method and portfolio variants. Drawdown summary and underwater curves. Market regime labels. Realized volatility. New surfaces ship monthly; subscriber roadmap drives priority.
Quant features, productized
The first commercially-available ML feature store for crypto. Model-ready feature columns derived off our underlying tape — regime tags, momentum cohorts, volatility regimes, order-flow imbalance, funding-curve shape factors, microstructure features. Built originally for internal research; opened to subscribers today. 26 signal endpoints live across 7 signal families; catalog continues to expand across 2026.
Bitcoin since 2009, Ethereum since 2015
Block-level aggregates, network economics, mempool state, exchange flows and reserves, miner-reward (coinbase) flows, smart-contract creation tracking, and the day's largest transactions. Plus engineered feature columns — NVT ratio, network congestion, supply velocity, address velocity, whale activity.
Cross-venue funding, US macro
Perpetual funding rates from Binance, Bybit, and OKX, plus a cross-venue aligned series for basis arbitrage and fade trades. US Treasury par yields from 1-month to 30-year tenor, TIPS real yields, and the weekly CFTC Commitment of Traders report — the macro tape joined directly to the crypto series.
The ML feature store.
Quant funds spend years engineering the feature columns that ride underneath their models — regime tags, momentum cohorts, microstructure signals, cross-asset stress indicators. We've built that layer for crypto, reconciled it against our tape, and made it available at the same flat subscription. There's no comparable product at this price point — and in most cases, no comparable product at any price point. The categories below are live or rolling out through 2026.
Volatility Regimes
HMM-tagged regime states per asset (calm / trending / blow-off), forward-look transition probabilities, regime-conditional return distributions.
Cross-Sectional Momentum
Lookback z-scores at 5/20/60-day horizons, decile cohort tags across the spot universe, momentum-quality (Sharpe-adjusted) variants.
Order-Flow Imbalance
Buy-vs-sell volume imbalance, top-of-book persistence, microstructure features off the transactional summary layer. BTC/ETH at hourly cadence.
Funding Curve Shape
Term-structure parameters off perpetual funding plus quarterly futures — level, slope, and convexity factors that anticipate funding regime shifts.
Correlation Regimes
Rolling block structure off the pairwise correlation grid. Detects when "crypto trades as one" versus when dispersion is back. Beta-to-BTC drift.
Liquidity Microstructure
Effective spread proxies, depth-adjusted volume, large-trade frequency, tradecount-to-volume ratios. Pre-computed for the high-volume pairs.
On-chain Network Stress
Fee-percentile regimes, hashrate-difficulty divergence, miner revenue compression, smart-contract creation acceleration. Derived from the on-chain stack.
Options-Implied Sentiment
Put/call skew, term-structure of IV, gamma exposure proxies, large-print direction. Built off the Greeks tape and the day's top-100 transactions.
Macro Cross-Asset Soon
BTC beta to risk assets, regime tagging conditioned on the yield curve, correlation shifts during macro tape stress windows. Q3 2026.
The full feature catalog is live at api.cryptodatadownload.com — schema, parameters, and field reference updated continuously as new signals ship. If you're using a feature in a research pipeline and want a specific one added or modified, email support@cryptodatadownload.com — the roadmap responds to subscriber requests directly.
Binance
OHLCV across spot, USDT-M and COIN-M futures, options chains, and tick granularity for BTC/ETH. A deep summary layer covering basis, options OI by maturity, volume profile, large trades, and order-flow imbalance.
GET Binance Spot Data
OHLCV bars for Binance spot symbols at daily, hourly, or minute granularity — selectable via the interval parameter (1d, 1h, 1m).
The canonical price-volume history used by virtually every consuming chart, backtest, or signal. Returns gapless OHLCV plus a volume_from row that resolves base-currency volume into quote-currency (USDT). The two marketorder_* columns isolate aggressive (market-order) flow — useful for distinguishing taker pressure from passive liquidity.
GET Binance Futures UM Data
OHLCV for Binance USDT-margined (UM) perpetual + dated futures, selectable interval (1d, 1h, 4h, 1m).
USDT-margined contracts settle in stablecoin and are the dominant Binance derivatives venue by volume. Use for futures backtesting, basis calculation, or cross-listing studies.
GET Binance Futures CM Data
OHLCV for Binance coin-margined (CM) inverse perpetual + dated futures.
Coin-margined contracts settle in the base asset (BTC, ETH, etc.) and historically attract long-term hedgers + miners. Useful for analyzing carry differentials vs UM and dated-contract basis.
GET Binance Options Data
OHLCV option chains for Binance options — query by date, underlying, specific instrument, or maturity.
Binance options are USD-settled European-style. Use this for backtest-grade option premium history.
GET Binance Tick Data
High-frequency aggregated tick bars for Binance — unix_open / unix_close window with per-window OHLC and trade counts.
The closest thing to raw tape we expose. Use for microstructure analysis, queue dynamics, or sub-minute volume profile.
GET Spot and Futures Prices & Corresponding Basis
Aligned daily spot and futures OHLCV plus the resulting cross-market basis (spot − futures, or annualized carry for dated contracts).
The simplest cash-and-carry view: compare what you'd pay in spot vs the embedded forward expectation. Sustained positive basis = contango; sustained negative = backwardation.
GET Binance Futures UM Summary Metrics
Daily UM derivatives microstructure metrics — sum / count of open interest, top-trader long-short ratio, taker buy/sell volume ratio.
The set Binance publishes for derivatives analytics — useful as positioning + sentiment proxies. toptrader_long_short_ratio is the canonical positioning skew; taker_long_short_vol_ratio is the flow skew.
GET Binance Options OI Totals
Daily total open interest aggregated across all expiries — calls and puts separately, per underlying.
The widest-angle OI view. Use to track aggregate options growth on Binance and gauge OI-implied positioning shifts.
GET Binance Options OI Totals by Instrument
Per-instrument OI snapshot — start_OI vs end_OI for each option, with day-over-day delta plus volume and greeks.
The most granular OI view we expose. Use to attribute aggregate OI shifts to specific strikes / expiries, and to identify newly-built positions.
GET Binance Options OI Totals by Maturity
Daily OI totals broken down by expiry — calls and puts separately.
Term-structure view of options positioning. Use to spot expiry-clustering (e.g. heavy front-month) and shifts in the OI curve.
GET Binance Flow Imbalance (Order Flow Delta)
Daily order-flow imbalance per symbol — delta between aggressive buy and sell volume, with raw and z-scored variants.
Net taker delta is a leading indicator for short-horizon directional moves. Z-scored against the symbol's own history makes the signal comparable across symbols.
GET Binance Large Trades (Whale Activity)
Individual large trades on Binance spot — every print over the symbol's whale threshold, with side, price, quantity, and notional.
The unfiltered whale tape. Use for event-driven studies (e.g. how often does a >$5M print precede a regime change?) or as a candidate signal for elephant-tracking strategies.
GET Binance Spot Volume Summary (S3)
Same daily spot trade-summary as the legacy endpoint but served from the S3-backed reader — supports a wider symbol universe and longer history.
Functionally equivalent to the legacy daily summary; use this whenever you need history beyond ~2 years or symbols outside the canonical 8.
GET Binance Daily Trade Summary
Per-day Binance spot trade-summary at the symbol level — VWAP, buy/sell counts and volumes, average trade size.
Use for cross-sectional venue comparison or as a daily input to flow-based signals.
GET Binance Hourly Trade Summary
Per-hour Binance trade-summary aggregations — same shape as the daily summary but bucketed by hour.
Use for intraday flow studies, hour-of-day seasonality, and event-window aggregations.
GET Binance Volume Profile (Price Distribution)
Daily volume profile per symbol — volume traded at each price bucket, with POC, VAH, and VAL.
The classic auction-theory view of where the day's volume settled. POC = price of highest volume; VAH/VAL bracket the 70% Value Area. Use as horizontal support/resistance candidates.
Deribit
The full Deribit derivatives stack — futures, perpetual funding, DVOL volatility indices, option Greek summaries by instrument and maturity, put/call ratios, large-trade detection, and the day's top transactions by USD notional.
GET DeriBit Futures Daily Data
OHLCV for Deribit dated and perpetual futures.
Deribit's futures are inverse-settled (in coin), making them the natural reference series for coin-margined basis trades and futures-vs-spot studies.
GET DeriBit Funding Data
Deribit perpetual funding-rate history — 8-hour and 1-hour decay components, index and previous-index prices.
Deribit funding decomposes into the 8h base and a 1h decay. Use the 8h column as the canonical funding observation; the 1h column is helpful for intra-window dynamics.
GET Deribit Options Timeseries
Same as the single-instrument timeseries endpoint, kept under the newer canonical path. Returns OHLCV history for one Deribit option.
Prefer this newer endpoint for new integrations.
GET DeriBit Volatility Data
OHLCV history of Deribit's DVOL volatility index — the exchange's equivalent of the VIX, computed from listed Deribit options.
DVOL is the cleanest market-implied volatility benchmark for BTC and ETH. Use as a regime indicator (high DVOL = stress / event pricing; low DVOL = calm) or as a hedging-cost proxy.
GET Deribit Options Flow Analysis
Net options flow per day — aggressive buyer vs aggressive seller direction, broken down by underlying.
Closest thing we have to a directional flow indicator on Deribit options. Positive net = aggressive buying / call-bid bias; negative = put-bid / hedging tone. Pair with the unusual-activity flags for higher-conviction signals.
GET DeriBit Greeks Summary by Instrument (S3)
S3-backed instrument-level greeks summary — per (instrument, date) with maturity, strike, putcall flag plus volume and premium.
Use whenever you need history older than the MySQL-backed endpoint can serve. Same fields, longer reach.
GET DeriBit Greeks Summary by Maturity (S3)
Greeks aggregated to the (underlying, maturity, putcall) level — useful for term-structure views of dealer positioning.
Reduces 100+ strikes-per-expiry into a single maturity row. Use for plotting term-structure heatmaps of net delta / gamma / vega.
GET Deribit Options Large Trades Analysis
Daily large-trade aggregations on Deribit options — count, total premium, and notional of prints over the dollar threshold.
Large trades capture institutional flow. Sudden spikes are the textbook unusual-activity signal that has driven monetization of products like Unusual Whales.
GET Deribit Options Put/Call Ratios
Daily put/call volume and OI ratios per underlying.
Classic positioning sentiment proxy. Put/call > 1 means more put activity; rising ratio = hedging interest or bearish positioning.
GET Top 100 DeriBit Option Transactions per Day in USD
Top 100 largest Deribit options transactions per (currency, date) ranked by USD premium.
Captures the day's headline prints — useful for noteworthy trade content and as input to whale-following strategies.
GET Deribit Options Volume Summary
Per-day volume summary on Deribit options — total volume, total notional in USD, and trade count per underlying.
Use as a normaliser for any flow signal (e.g. large-trade share = large_premium / total_premium).
GET Deribit Options Transactions
Raw individual transactions on Deribit options for the requested date and currency — every print with timestamp, instrument, price, size, side, and Greeks.
The unaggregated tape. Use for microstructure work, custom flow models, or to build downstream signals not exposed elsewhere.
Funding Rates
Perpetual funding rates from Binance, Bybit, and OKX, plus a cross-venue aligned series so you can compare basis decay and fade arbitrage across exchanges in a single query. (Deribit perpetual funding is grouped with the Deribit derivatives stack above.)
GET Aligned Cross-Venue Funding Rates
All three venues' funding rates joined on 8h settlement buckets (00:00 / 08:00 / 16:00 UTC). Returns side-by-side venue columns plus median, max spread, and pairwise spread columns.
One-call cross-venue arbitrage view. Look at max_spread per bucket to spot dislocations; the pairwise columns tell you which pair is worth trading. NaN columns indicate a venue with no settlement at that bucket.
GET Binance Funding Rates
Raw 8h-cadence funding rate events for Binance USD-M perpetual contracts. Three rows per day (settlement at 00:00 / 08:00 / 16:00 UTC).
The actual paid-or-received funding rate at every 8h settlement. Sum across the day for daily P&L on a perp position; combine with the aligned endpoint to spot cross-venue arbitrage.
GET Bybit Funding Rates
Raw 8h-cadence funding rate events for Bybit USDT-margined linear perps. Same schema as the Binance endpoint.
Bybit's actual funding rate per settlement. Combine with Binance/OKX endpoints (or use /funding/aligned/) to spot cross-venue dislocations and basis trades.
GET OKX Funding Rates
Raw 8h-cadence funding rate events for OKX USDT-margined SWAP contracts. Uses OKX's realizedRate field (actual paid) over fundingRate (predicted).
OKX's actual funding. The venue_symbol column preserves OKX's native instId (e.g. BTC-USDT-SWAP) for debugging or cross-reference, while symbol is the normalized form for cross-venue joins.
Options Analytics
Volatility surfaces, term structures, IV vs RV spreads, higher-order Greeks (charm, vanna), pin candidates, and a Black-Scholes-Merton pricing endpoint — the analytics layer most retail platforms charge five figures a year for.
GET Higher-Order Greeks
Per-instrument vanna, charm, vomma, and speed for every listed Deribit option in the requested base currency, computed via scipy-free BSM central finite differences.
Sensitivities that matter for dealer hedging beyond first-order greeks. Vanna = dDelta/dVol (skew exposure). Charm = dDelta/dTime (decay-adjusted delta hedging). Vomma = dVega/dVol (convexity in vol). Speed = dGamma/dSpot (gamma curvature).
GET Pin Candidate Strikes
Successor to the classic max-pain heuristic. Ranks Deribit option strikes by a composite of open interest, traded volume, and absolute gamma. Returns top 50 per (base_currency, date) with within-expiry and overall ranks.
Where the gamma magnets are. Strikes with high pin scores tend to act as price attractors into expiry, especially within 0–3 DTE. Rank 1 within near-DTE expiries is the marquee pin level traders watch.
GET Options Term Structure
At-the-money implied volatility for every listed Deribit options maturity for the requested base currency and date, plus a contango/backwardation flag and the % deviation vs the 30-day ATM IV.
The shape of the IV term structure. Front-month rich vs back = near-term event pricing; back-month rich = forward stress or structural hedge demand. contango_flag=1 (downward slope) usually = calm; =0 (backwardation) = stress.
GET Options VWAP per Strike-Day
Volume-weighted average price for each Deribit option strike × maturity × putcall combination traded on the requested date, with separate VWAPs in the base currency (BTC/ETH) and USD, plus volume and mean IV.
The actual execution-grade option prices — what traders paid on average. Useful for portfolio mark-to-market, backtest fills, and identifying which strikes saw the most flow on a given day.
GET Volatility Cone
Per-symbol realized volatility percentiles across {5, 10, 20, 30, 60, 90}-day windows and 4 RV estimators (close-to-close, Parkinson, Garman-Klass, Yang-Zhang), with each window's current RV positioned against its 2-year historical distribution.
"Where are we in the vol cone today?" When percentile_today is low across all windows, vol is suppressed relative to history (potential mean-revert higher); when high, vol is elevated (potential exhaustion). Trade vol cones by buying/selling vol at the extremes.
GET IV vs RV Spread
Variance-risk-premium proxy: ATM 30-day implied vol minus 30-day realized vol (4 methods), with the spread expressed in % and a variance-risk-premium percentage. Also includes a rolling 30d average for context.
When iv_rv_spread_pct is positive, implied vol is rich vs realized (good for vol sellers); when negative, implied is cheap (vol buyers' market). The 30d avg shows whether today is anomalous vs the recent regime.
GET Implied Volatility Surface
Daily IV surface grid — 9 moneyness × ~13 maturities — smoothed via Nadaraya-Watson kernel regression. Each row is one (strike, DTE, moneyness) cell with both the raw IV and the kernel-smoothed value.
The full implied vol topology — readable as a 2D heatmap by strike × maturity. Useful for spotting term-structure waves, skew steepening at specific DTEs, and discrete listings (single-event days). n_observations indicates how many quotes contributed to each cell — sparse cells have higher smoothing weight.
GET Black Scholes Merton (BSM) Option Pricing Model
On-demand BSM option-pricing calculator — returns theoretical price plus full first- and second-order greeks for the provided parameters.
Use for ad-hoc valuation, sensitivity studies, or to cross-check vendor quotes.
On-chain Data Features
Bitcoin since 2009, Ethereum since 2015. Block-level aggregates, network economics, mempool state, exchange flows and reserves, the day's largest transactions, plus engineered on-chain feature columns — NVT ratio, network congestion, supply velocity, whale activity.
GET Address Velocity
Daily active addresses on the chain with 7d/30d rolling averages, velocity ratios at multiple horizons, and a 90-day z-score. v1 is ETH-only — BTC requires unique-input/output address ingestion (deferred to Phase 2).
Address velocity is a network-usage proxy. Rising active addresses + rising prices = healthy adoption; rising addresses + flat prices = utility usage without speculation; falling addresses = waning engagement.
GET ETH Contract Growth
Daily count of newly-deployed Ethereum smart contracts with 7d/30d/90d rolling averages, 1-year z-score, and a categorical growth label.
Contract growth is a leading indicator of developer activity. Bursts often correlate with new protocol launches, airdrops, or NFT cycles. Use as a forward-looking developer-engagement signal.
GET Network Congestion Score
Cross-chain congestion composite that blends BTC mempool congestion, BTC fast-fee z-score, ETH fast-gas z-score, and ETH block utilization into a single score and categorical label.
A unified "how full is the network" view across BTC + ETH. congestion_label=extreme flags days when both chains are heavily loaded — usually market-event days. Useful as a macro overlay or as input to a regime composite.
GET NVT Ratio
Network Value to Transactions ratio per chain — market cap divided by on-chain transaction volume in USD. Returns raw NVT, 14-day smoothed NVT, and 180-day z-score.
The original "crypto P/E" — when NVT is high, the network is valued highly relative to economic throughput (overvalued/speculative); when low, undervalued relative to usage. Z-score normalizes for regime.
GET Supply Velocity
How fast the circulating supply is changing hands on-chain. Velocity is the daily on-chain volume divided by circulating supply, with rolling averages, a 180-day z-score, and a categorical regime label.
Rising velocity = coins moving more, often around catalysts/distribution. Falling velocity = HODL behavior, supply parking. Z-score normalizes for regime; label categorizes for ML use.
GET Whale Activity
Daily share of on-chain transaction volume from large transactions (whale flows) per chain, with the 90-day z-score and a categorical whale flag.
whale_flag flips to elevated or extreme when today's large-tx share is statistically unusual. Often precedes price moves (smart-money positioning ahead of news/catalysts). Track per-chain to compare BTC vs ETH whale behavior.
GET BTC Mempool State
Daily snapshot of the Bitcoin mempool from mempool.space: median fee (sat/vB), fast/slow confirmation fee tiers, mempool size in vBytes, pending block count, and a normalized congestion percentage.
congestion_pct=100 means the mempool is fully saturated — expect slow confirmations and elevated fees. Watch this around major events to anticipate fee spikes. The fee tiers tell users what to pay for fast/slow inclusion.
GET ETH Exchange Flows
Daily net flow per exchange, derived from the day-over-day balance delta across labeled addresses. Returns total reserves both days, net flow in ETH and USD, addresses observed, and a status flag.
Positive = balance INCREASED at the exchange's labeled addresses (net inflow). Negative = balance DECREASED (net outflow). Large inflows often precede sell pressure; large outflows = accumulation or self-custody migration. Status fields flag cold_start (first day of coverage) and stale (balances unchanged — possible RPC failure).
GET ETH Exchange Reserves
One row per labeled exchange wallet address with ETH balance (native and USD), role (hot/cold/deposit_aggregator), source attribution, and block number used for the snapshot. v1 universe: ~32 publicly-attested addresses across Binance/Coinbase/Kraken/Bybit/OKX.
The actual ETH stash at known exchange wallets, attributed to each address. Sum across rows for total exchange-held ETH. Use individual addresses to track specific cold wallet movements.
GET ETH Gas Prices
Daily ETH gas-price snapshot from Etherscan V2 gas oracle (with Owlracle fallback): average / slow / fast gas in gwei, base fee, and priority fee.
Read avg_gas_gwei for the typical user experience that day. base_fee_gwei is the burn floor; priority_fee_gwei is the validator tip on top. Spikes here correlate with on-chain activity (DEX volumes, NFT mints, network stress).
GET ETH Transaction Aggregates
Daily Ethereum transaction-level aggregates from AWS Public Blockchain: transaction count, total ETH value moved, gas used, EIP-1559 ETH burned, unique sender/receiver address counts, and failed-tx count.
Network heartbeat. Tx count + unique senders gauges activity; gas_burned_eth shows deflationary pressure (post-Merge); failed_tx_count signals network stress (high during NFT mints, oracle congestion).
GET BTC Market Capitalization
BTC market cap timeseries — circulating supply × close price, sourced from blockchain.com.
The denominator of MVRV / NVT-style ratios. Self-contained for back-of-envelope sizing.
GET BCHAIN Aggregated Data
Daily-aggregated blockchain.com series — transactions per day, mining difficulty, hash rate, mempool size, etc., per the BCHAIN code.
The historical macro on-chain reference. Use for long-horizon studies (multi-year hash-rate / difficulty cycles, halving epochs).
GET Blockchain Summary Metrics
Daily summary metrics for the BTC blockchain — transaction count, fee totals, miner revenue, average tx size, etc.
A header-level dashboard of network activity. Pair with the per-block / per-tx endpoints for drill-down.
GET On-Chain Ethereum (ETH) Summary Contract Creation Metrics
Daily aggregate of ETH smart-contract deployments — total contracts, unique deployers, gas burned.
Macro view of ecosystem build-out velocity. Spikes track new DeFi cycles, NFT trends, or scam-token bursts.
GET Blockchain Transaction Metrics
Daily BTC transaction-level aggregates — input/output sums, top transactions, fee distribution, and aggregate value moved.
Use this to track exchange-scale flows or to derive market-cap-like statistics from on-chain throughput.
GET BTC Coin Creation(s)
Per-day count of newly created BTC (coinbase rewards minted) and the per-block subsidy.
Use to track halving cycles and chart issuance vs adoption.
GET Top 50 BTC Transactions per Day
Daily ranked list of the 50 largest on-chain BTC transactions by output value.
The classic whale-watching feed. Pair with public attribution to identify exchange transfers or treasury moves.
Risk & Volatility
Correlation grids and rolling correlations, drawdown summary and underwater curve, regime label, VaR by method, portfolio-level VaR, standalone VaR across 1,100+ symbols, and realized volatility. A growing surface library; new endpoints shipping monthly.
GET Trading Correlations
Pairwise rolling correlations between symbol pairs over the requested window and calculation method.
Use to construct correlation matrices for portfolio construction or to spot regime shifts (when historically uncorrelated pairs start co-moving).
GET Rolling Correlations
Rolling pairwise Pearson correlations across the top-30 crypto universe — 435 unique (symbol_a, symbol_b) pairs at the requested window size.
When correlations broaden, idiosyncratic alpha shrinks (single-name picks blend into beta). When correlations break down, dispersion strategies and pair trades come alive. Watch the cross-section of correlations to a single name (e.g. BTC) for risk-on/risk-off rotation.
GET Drawdown Summary
Drawdown summary for the requested symbol — all-time max drawdown %, current drawdown vs running peak, peak/trough dates, peak-to-trough duration, recovery time (peak-to-recovery), and an in-drawdown flag.
The total downside experienced by the symbol over its full history. Useful for position-sizing, risk budgeting, and stress-testing. current_drawdown_pct tells you how much we're underwater right now vs the all-time high.
GET Underwater Curve
The complete daily underwater curve for the symbol — % drawdown from the running peak at every historical date, suitable for plotting an underwater chart.
Visual history of every drawdown the symbol has experienced. Plot this to see how long the asset spent below its prior peak and how deep each drawdown went. Sibling of the /drawdown/ summary.
GET Market Regime Label
Daily composite regime label across the cross-asset universe, derived from 5 z-scored components (breadth, vol, funding, COT positioning, IV). Returns a continuous regime_score in [−1, +1] plus a categorical label.
A single-number summary of the market's risk appetite. risk_on (score > 0.3) = broad participation, suppressed vol, neutral-to-positive funding; risk_off (< −0.3) = compressed breadth, vol spike, negative funding extremes. neutral in between.
GET Value at Risk (VaR)
Per-symbol Value-at-Risk and Conditional VaR (Expected Shortfall) at the requested method (historical sim or parametric/Cornish-Fisher), expressed as a negative percentage return. Multiple confidence levels (90, 95, 99) and windows (30, 60, 90, 252) returned in one call.
The worst expected loss for the symbol at the given confidence. var_pct = -0.022 at 95% confidence means there's a 5% chance of losing more than 2.2% on a single day. CVaR is the average loss in the tail beyond VaR.
GET Portfolio Value at Risk
Portfolio-level Value-at-Risk for one of two pre-built static portfolios (cap_weighted_top10, eq_weighted_btc_eth_sol), including portfolio VaR, per-symbol component VaR (decomposition of total risk), per-symbol marginal VaR (how much VaR changes per unit weight), and the symbol/weight arrays.
Which holdings drive your portfolio risk. Component VaR adds up to portfolio VaR; marginal VaR tells you which symbol to trim or add to most efficiently reduce overall risk.
GET Value at Risk by Symbol
Worst-case observed loss per symbol with the date it occurred and the methodology used (parametric vs historical).
A point estimate of tail risk — useful as a sanity-check upper bound for risk budgets. For richer VaR shapes (multiple confidence levels, rolling windows), use the expansion VaR endpoint.
GET Realized Volatility
Daily realized vol estimates for the requested symbol across 4 estimators (close-to-close, Parkinson, Garman-Klass, Yang-Zhang) and 6 lookback windows (5, 10, 20, 30, 60, 90 days). 296 symbols covered.
The historical volatility surface for backtest features, vol-targeting, and vol-cone comparison. CC is naive; Parkinson uses high-low (more efficient); GK uses OHLC; YZ handles overnight gaps. Compare against ATM IV to gauge variance-risk premium.
Market Internals
Equity-market analytics adapted for crypto — daily new 52-week highs and lows counts, an advance-decline line, the share of pairs trading above key moving averages, and a general-purpose technical-indicator surface.
GET Daily New 52 Week Highs and Lows Counts
Daily count of symbols making new 52-week highs vs new 52-week lows across the supported Binance universe.
The classic breadth indicator. Persistent high count of new highs = broad-based bull regime; high count of new lows = broad weakness. Composite (highs − lows) often leads index price.
GET Advance/Decline Calculations
Daily advance/decline counts and net A/D across the supported universe.
Sister breadth indicator to the % above MA series — counts how many symbols closed up vs down. Net A/D (advances − declines) is the textbook breadth heartbeat.
GET Pair Counts Trading Above Moving Averages
Daily counts and percentages of supported pairs trading above MA50, MA100, and MA200.
Standard market-breadth measure. % above MA200 below 30 historically marks deep-oversold; above 80 marks broad-momentum / euphoria.
GET Technical Analysis Indicators
Per-symbol daily OHLCV with a battery of TA-Lib indicators pre-computed — ADX, RSI, MACD, Bollinger Bands, Stochastic, ATR, and others.
Use as a ready-to-model technicals panel. Saves a roundtrip to TA-Lib in client code; values are computed on the same OHLCV history exposed by the Binance Spot Data endpoint.
ML Feature Store
Pre-computed, model-ready feature columns derived off the underlying tape. Subgrouped below by family. See the flagship section above for the conceptual framing — the rows here are the live endpoint paths with their full descriptions and field schemas.
Cross-Sectional · 5GET Beta vs BTC
Per-symbol rolling beta to BTCUSDT at 30/60/90/180 day windows, plus 30-day idiosyncratic return and 60-day R-squared. BTCUSDT itself is excluded.
How much of the symbol's daily move can be attributed to BTC. Beta > 1 = amplifies BTC moves; < 1 = dampens; idio_return isolates the symbol's own alpha. High R² means BTC explains most variance; low R² leaves room for non-beta factors.
GET Momentum Factor
Classical academic momentum factors: returns over the past 12/6/3 months excluding the most recent month, plus their z-scores against the cross-section.
Jegadeesh-Titman-style momentum — buy past winners, sell past losers. The skip-last-month design avoids short-term mean reversion. Positive z-scores = strong momentum vs peers.
GET Return Rank
Where the symbol ranks against the top-30 universe on trailing returns at five horizons (1d/5d/20d/60d/180d). Returns both the cross-sectional z-score and the integer rank for each horizon.
Classic cross-sectional momentum factor inputs. Rank 1 = best performer; rank 30 = worst. Z-score answers "how many standard deviations above/below the universe mean". Combine ranks across horizons for momentum-quality signals.
GET Reversal & Hurst Regime
Short-horizon mean-reversion score (1d, 5d) plus a 60-day Hurst exponent and a categorical regime flag indicating trending vs mean-reverting behavior.
When regime_flag is mean_reverting (Hurst < 0.5), pullbacks tend to revert; when trending (Hurst > 0.5), trends extend. The reversal scores then have different alpha properties — use them in trend regimes for short-term scalps, avoid them in mean-revert regimes.
GET Volatility Rank
Where the symbol sits in the universe on 30-day realized vol — z-score and rank.
Identifies vol-leaders vs vol-laggards. High vol_rank = unusually volatile vs peers (potential breakout or news catalyst); low rank = quiet vs peers (compression, potential breakout setup).
GET Large-Trade Share
Fraction of daily notional that came from "large" trades, plus 7d/30d rolling averages and the 1-year percentile.
When large_trade_share is high vs its 1y percentile, institutional/whale flow is dominating that day. Low share = retail-driven tape. Use as a proxy for "smart money" participation.
GET Order-Flow Imbalance
Daily order-flow imbalance for the symbol — buy volume share, sell volume share, raw OFI (buy − sell), and the 60-day z-score of OFI.
Positive OFI z-scores indicate aggressive buying pressure (taker-side buy dominance); negative = sell pressure. Extreme z-scores often precede short-term price continuation in the OFI direction.
GET Volume Profile
Daily volume profile statistics: Point of Control (highest-volume price), Value Area High and Low (boundaries of the value area containing ~70% of volume), and the value-area percentage.
POC = fair value benchmark for the session. VAH/VAL bracket the consensus pricing range; trades outside the VA tend to revert. Useful for intraday range estimation and breakout confirmation.
GET Cross-Exchange Spread
Daily price spread per pair across two venues — typically Binance (USDT) vs a USD-quoted venue (Bitstamp, Coinbase, etc.). Returns close prices, spread %, and 30-day rolling average.
Persistent spreads indicate funding/FX/liquidity dislocations between venues. Mean-reverting spread crossings are basis-trade opportunities. The 30d avg gauges what's "normal" for this pair.
GET Implied Move
Expected price move (% of spot) implied by the ATM straddle at three target maturities (7d / 30d / 90d). For each target DTE, also returns the actual available maturity and source strike used.
What the options market expects the price to move (in either direction) over the next 1 week / 1 month / 3 months at ~68% confidence (one stdev). Use to size positions, set realistic price targets, or quote vol products.
GET Open-Interest Concentration
Open-interest concentration metrics per maturity: Herfindahl-Hirschman Index, the top-3 strikes' share of total OI, and the strike with the most OI.
High HHI / high top_3_share = concentrated positioning at a few strikes (pin candidates, potential gamma squeeze setups). Low values = broad participation across strikes. max_strike tells you THE level the market cares about.
GET Skew Signal
25-delta risk reversal per Deribit maturity (call IV − put IV at 25Δ), with 7-day and 30-day rolling averages, plus a categorical skew label per maturity.
rr_25d measures put-vs-call asymmetry. Negative values (puts richer than calls) = bearish skew, market pricing downside protection demand. The 7d/30d avgs show whether today is an extension or a fade. skew_label categorizes for ML pipelines.
GET Smile Curvature
Per-maturity IV smile descriptors: total smile width (wings minus ATM), wing asymmetry (put wing − call wing), and a kurtosis proxy capturing how peaked vs flat the smile is.
Wide smiles + high kurtosis = market pricing rare-event risk; flat smiles = consensus on the future. Positive wing_diff = downside protection bid more than upside calls (classic crypto bear-skew); negative = call demand.
GET Anomaly Score
Per-symbol multivariate anomaly score using Mahalanobis distance from the rolling 90-day distribution of 5 features (log_return_1d, rv_30d, vol_ratio, funding_z, oi_change_5d). Higher distance = more anomalous today.
When anomaly_flag flips to extreme, today is statistically unusual across the full multivariate factor set — vol regime + return + positioning are jointly out-of-distribution. Worth investigating; often coincides with news/event days.
GET Cluster Membership
Universe-wide cluster assignments from K=6 hierarchical agglomerative clustering on the rolling 60-day correlation matrix. Returns one row per symbol with cluster_id, cluster_size, average intra-cluster correlation, and distance to cluster centroid.
Which crypto assets are moving together right now. Symbols in the same cluster have similar return profiles over the last 60 days. Use for diversification (rotate across clusters) or pair trades (within-cluster mean reversion).
GET Regime State
Per-symbol posterior probabilities across 3 regimes — trending up, trending down, and mean-reverting — from an EM-fit Gaussian mixture model. Plus a regime-transition flag when today's MAP regime differs from yesterday's.
Use regime probabilities to gate strategy execution. Trend-following on trending_* regimes, mean-revert on mean_reverting. regime_transition_flag=1 is a hint that the previous-day's strategy assumption may have invalidated.
GET Funding Regime
Per-perpetual funding-rate regime: current funding, 7d/30d/90d averages, consecutive days of positive/negative funding, 1-year percentile, and a categorical regime label (e.g. extreme_negative, elevated_positive).
Funding regime tells you crowdedness. extreme_negative percentiles + sustained negative days = aggressive shorts (potential squeeze setup). elevated_positive = aggressive longs (potential mean-revert).
GET Open-Interest Change
Open-interest deltas at three horizons, paired with price change and categorical flags identifying whether the change reflects long_buildup, short_buildup, long_squeeze, or short_squeeze patterns.
OI rising + price rising = long buildup (trend continuation likely). OI rising + price falling = short buildup. OI falling + price rising = short squeeze. OI falling + price falling = long capitulation. Position-shift flags compress this 4-corner logic into a tag.
GET Perp Basis Time-Series
Time-series-decomposed perp basis: current basis %, EWMA-smoothed values at 3d/7d/30d half-lives, and annualized carry.
The current basis vs its different EWMA half-lives shows regime persistence. When current << ewma_30d, basis is collapsing fast (often = unwind). Annualized carry is what you'd earn (or pay) running the basis trade.
GET COT Positioning
CFTC Commitment of Traders data for the requested category: money-manager net positioning, total open interest, net as % of OI, 5-year z-score, and categorical regime label.
Where the smart-money positioning sits. Extreme z-scores (very_long / very_short) often precede mean-revert moves. The 7 categories cover the primary crypto-related futures: BTC, ETH, micro contracts, plus gold, USD, and SPX as macro overlay.
GET Real-Rate Regime
10-year real yield (TIPS-based) with 5d/30d/90d deltas, 5-year z-score, and a categorical regime label (very_low / low / neutral / high / very_high).
Real rates are the cleanest macro driver of crypto valuations. Falling real rates (z low) historically = easier financial conditions, supportive for risk assets. Rising real rates (z high) = tighter conditions, headwind for crypto.
GET Yield-Curve PCA
Classic 3-factor decomposition of the US Treasury yield curve. PC1 explains parallel level shifts, PC2 captures slope (steepening/flattening), and PC3 captures curvature (humps/inversions).
Distills the full yield curve into 3 numbers. PC1 = rates regime; PC2 = recession signal (PC2 falling = curve flattening); PC3 = monetary-policy distortions. Var explained columns show how much each factor matters today.
GET Futures Basis Term Structure
Daily basis (futures vs spot) for the requested base currency across the perp and quarterly contracts on Binance. Includes annualized carry % and days-to-expiry for each contract.
The cost/yield of being long/short futures vs spot at each maturity. Positive basis = contango (futures > spot); negative = backwardation. The annualized carry is the basis-trade APY before funding/borrow.
GET Macro Yield Spreads
Treasury curve slopes and real-nominal spreads computed from the US Treasury par yield curve: 2s10s, 5s30s, 3m10y, plus real-nominal at the 5y and 10y points.
Traditional macro signals overlayed on crypto: inverted 2s10s historically precedes recessions; widening 3m10y signals growth optimism; real-nominal spreads indicate inflation expectations.
GET Spot Returns
Per-symbol forward-looking and backward-looking returns at 1d, 5d, 20d, 60d horizons in both log and simple forms. 296 symbols covered.
The building block for momentum, mean-reversion, and beta features. Choose log returns for additivity (e.g. summing daily returns to monthly) and simple for cumulative wealth calculations.
GET Spot VWAP
Daily volume-weighted average price for the symbol, derived from the trade_summary silver. Constrained to the 9 symbols that have trade_summary partitions — broader spot universe is via the OHLCV endpoints.
True average execution price for the day, weighted by traded volume. Better than (high+low+close)/3 or close for fair-value benchmarks, mark-to-market, and back-fill validation.
Aggregates & Dashboards
Composite endpoints that pre-join the most-used surfaces into a single response — the ML feature blend, on-chain dashboard, options dashboard, and per-symbol risk panel. Built for dashboards that don't want to make twelve calls.
GET All ML Features Composite
Wide-row composite of all 22 Phase-2 ML features joined on the (symbol, date) key. ~70 columns covering return / vol / momentum / microstructure / options-implied / positioning / macro / regime dimensions in one row.
The ready-to-model feature vector for crypto cross-sectional and time-series strategies. One API call replaces 22 round-trips during model input prep. Pass fields= to return only a subset of columns (saves bandwidth on large training pulls).
GET On-Chain Dashboard Composite
One-row composite view of network state and derived signals for BTC or ETH — mempool/gas, congestion composite, whale activity share, NVT, supply velocity, contract growth.
The 'on-chain headline' panel for a Pro UI. Pairs naturally with the Risk Panel and Options Dashboard for a three-quadrant market overview.
GET Options Dashboard Composite
One-row headline view of the day's volatility surface — ATM IV, IV-RV spread, term-structure contango flag, 25Δ skew, top pin strike, and DVOL — for the requested base currency.
The 'what's BTC vol doing today' panel. Designed to back a Pro-tier dashboard or pre-trade vol gut-check. Composes 6 upstream parquets that would otherwise require 6 separate API calls.
GET Per-Symbol Risk Panel
One row per (symbol, date) summarizing the full risk view: realized vol (4 methods), ATM IV, IV-RV spread, drawdown depth + max, VaR at 95/99 confidence, and the cross-asset regime label.
The per-symbol risk header for a Pro UI table. One call gives you everything you'd display in a 'risk-at-a-glance' row.
US Government
Mirrored from US Treasury and CFTC so you can join macro tape against the crypto series without leaving the API. Treasury yields publish daily; COT publishes Fridays.
GET Committment of Traders (COT) Timeseries
Weekly CFTC Commitments-of-Traders positioning data — long / short / spread positions by trader category for the requested contract.
The textbook futures-positioning dataset. Use the Money-Managed long-short delta as a sentiment / extreme-positioning indicator. New report every Friday for prior Tuesday.
GET Treasury Par Yield Curve Rates
Daily US Treasury par yield curve — nominal rates across the canonical maturity grid (1m, 2m, 3m, 6m, 1y, 2y, 3y, 5y, 7y, 10y, 20y, 30y).
The benchmark risk-free curve. Use as the discount curve for option pricing or to monitor curve dynamics (slope, level, twist).
GET Treasury Par Yield Curve Real Rates
Daily US Treasury TIPS (real) yield curve — inflation-adjusted yields at the canonical maturities (5y, 7y, 10y, 20y, 30y).
The market-implied real-rate expectation. Inflation-sensitive trades, real-rate-based regime filters, and gold/Bitcoin macro studies all key off this curve.
Discovery
Utility endpoints that tell you what's available — list valid symbols, available date ranges, contract names, and coverage windows — so your script can ask before it pulls.
GET Available COT Contract Names / Groups
JSON array of CFTC Commitments-of-Traders contract names / asset groups exposed by the COT timeseries endpoint.
The COT report covers ~120 contracts across crypto, equities, FX, commodities. This catalog returns the exact name strings the timeseries endpoint accepts.
GET Available Binance Symbols
JSON array of every Binance symbol across spot + USDT-perp + coin-perp + options (union view).
Use as the master Binance universe probe — answers 'is symbol X anywhere on Binance?' in one call. For per-product-line views, use the more specific endpoints (Available OHLC Binance Symbols, Available Binance Futures Funding Symbols).
GET Available DeriBit Futures OHLC Symbols
JSON array of every Deribit dated-futures contract symbol exposed by the futures OHLC endpoint.
Includes both currently-listed contracts and expired ones. Useful for term-structure history work where you need to reconstruct past quarterly curves.
Full data schema →GET Available DeriBit Funding Symbols
JSON array of every Deribit perpetual symbol with stored funding-rate history.
Deribit perp funding uses an 8h-base + 1h-decay structure unique to the venue. This catalog returns the set of perps we have stored history for.
Full data schema →GET Available Deribit Option Symbols
JSON array of every Deribit option instrument name with stored OHLC history.
The full options catalog — typically several hundred to a few thousand strikes across active expiries. Filter client-side by underlying / DTE / strike depending on use case.
GET Binance Trade Summary Available Dates
JSON array of dates with stored Binance daily spot trade-summary records.
Use as a coverage probe before iterating over a date range on the daily trade summary endpoint. Useful because the trade summary universe is narrower than the spot OHLC universe (~9 canonical symbols vs the full Binance universe).
GET DeriBit Greeks Available Dates
JSON array of dates with stored Deribit greeks summary records.
Use as a coverage probe before requesting the Deribit greeks summary for a date range. Greeks summary lags the OHLC catalog by ~1 session (computed end-of-day).
GET Deribit Options Transactions Available
JSON array of (currency, date) coverage objects indicating which dates the Deribit options transactions endpoint has stored prints for.
Use as a coverage probe before iterating over a date range — saves a 404 round-trip on dates with no transactions. The response is more structured than the plain symbol catalogs (one object per (currency, date) pair).
Quick start — five seconds in
Authenticate with a token — header or URL param. Pick a path, pass parameters, choose format=csv|json|xlsx. That is the whole interface.
# BTCUSDT daily candles, as CSV curl -H "Authorization: Token YOUR_KEY" \ "https://api.cryptodatadownload.com/v1/data/ohlc/binance/spot/" \ -d "symbol=BTCUSDT" \ -d "interval=1d" \ -d "format=csv" -G
import requests, pandas as pd r = requests.get( "https://api.cryptodatadownload.com/v1/data/ohlc/binance/spot/", headers={"Authorization": "Token YOUR_KEY"}, params={"symbol": "BTCUSDT", "interval": "1d", "format": "json"}, ) df = pd.DataFrame(r.json())
# Paste into any browser. Auth via URL param — instant CSV download. https://api.cryptodatadownload.com/v1/data/ohlc/binance/spot/?symbol=BTCUSDT&interval=1d&format=csv&auth_token=YOUR_KEY
You've seen the surface. Plug in.
Every endpoint, every format, the full feature store — at $79.99/month, flat. Cancel from your profile whenever; resume whenever. No volume meters, no surprise invoice.