Crypto Carry Scanner

Compare current USDT perpetual funding snapshots across Binance USD-M, Bybit Linear, Gate USDT Futures, and Bitget USDT Futures. The scanner normalizes each venue's reported funding interval, pairs only exact matching base symbols, applies your per-leg volume screen, subtracts your explicit annualized cost assumption, and ranks the remaining momentary estimates.

This is read-only informational research tooling. It does not recommend, prepare, route, or place trades.

What you get

Each successful run writes one dataset item containing:

• requested, successful, and failed venue status;
• warnings for malformed records, fallback interval use, pagination limits, or partial venue failure;
• the number of usable market records and matching opportunities;
• up to 50 ranked opportunity rows;
• both legs' symbols, venues, raw funding rates, funding intervals, annualized rates, reported 24-hour quote volume, and next funding timestamps when available;
• gross APR arithmetic, your annualized cost deduction, and the resulting net APR estimate;
• explicit flags confirming that the result is informational only and that no account, credentials, or orders were used.

A valid scan may return zero opportunities when at least two venues succeeded but no exact base pair passed the volume filter.

Input

All fields are optional and have bounded defaults:

• venues: two to four values from binance, bybit, gate, and bitget; default is all four.
• minVolumeUsd: minimum reported 24-hour quote turnover for each leg, from 0 to US$1 trillion; default US$10 million.
• annualizedCostPct: annualized cost percentage subtracted from every gross spread, from 0% to 100%; default 3.8933%.
• maxResults: 1–50 nested rows; default 20.
• timeoutSeconds: 1–15 seconds per request; default 10.
• retries: 0–2 retries after the first request; default 1.

Do not submit a URL, API key, secret, file, account, wallet, position, order, prompt, or customer identifier. Those fields are not accepted.

Data behavior

The scanner sends credential-free HTTPS GET requests only to fixed public market-data endpoints on:

• fapi.binance.com
• api.bybit.com
• api.gateio.ws
• api.bitget.com

Any failed venue is recorded as a warning when at least two other venues remain usable. Fewer than two usable venues ends the run without a result. Binance and Bybit may block requests from some Apify Cloud regions; Gate and Bitget provide additional fixed public sources, but no venue is guaranteed reachable from every region. Each response is capped at 25 MiB, and all response chunks across pages and retries share a 40 MiB download cap. An 85-second scan-work budget operates within the 90-second run limit. Exceeding any limit fails closed.

The result is one dataset item with at most 50 opportunity rows. Standby mode is not used, memory is capped at 256 MiB, and the Actor requires only limited input, fixed-host network, dataset, and billing permissions.

Pricing

The paid price is US$0.15 per valid result. In pay-per-event mode, the single dataset write and the single crypto-carry-scan-result event are combined. Invalid input, fewer than two usable venues, result-generation failure, and dataset-write failure do not produce a paid result.

If the run is offered in free mode, the same result is written without a paid event. The dataset result itself contains no payment, price, or charge metadata.

Privacy

Input is limited to public venue selections and bounded numeric assumptions. No credentials, account identifiers, positions, orders, files, or customer text are requested or used.

The selected venues and network providers can observe ordinary request metadata such as IP address, time, and user-agent. Apify can retain run input, logs, and dataset output under its plan and platform policies. No no-store behavior is promised.

Limitations and risk

The scanner uses current reported funding and turnover, not executable order-book depth. Annualizing a snapshot does not predict future funding or realized return. Funding can change or reverse before settlement, and venue settlement times or contract terms can differ.

The calculation omits or simplifies basis divergence, fee tiers, maker rebates, slippage, legging, margin, liquidation, collateral impairment, borrow, transfer delay and cost, outages, tax, jurisdiction, and venue insolvency. Losses can exceed the apparent funding income.

This output is not financial, investment, legal, or tax advice, is not a suitability assessment or trade recommendation, and does not guarantee profit or any return. Verify current markets, rules, depth, costs, collateral, and local restrictions independently before making a decision.