Skip to main content

Flat Files - Overview

Welcome to the CoinAPI Flat Files documentation. This API provides access to historical cryptocurrency market data in flat file format, allowing for efficient retrieval and analysis of large datasets.

What is the Flat Files?

The Flat Files is a powerful tool for accessing comprehensive historical market data. It currently offers access via an S3-compatible API, allowing you to use familiar S3 operations and tools to retrieve data.

Available Data Types

Our Flat Files provides access to the following types of data:

  1. Quotes: Best bid and ask prices with associated volumes.
  2. Trades: Executed trades with price, volume, and direction.
  3. Limit Book Data: Full order book snapshots and updates.

Comming soon:

  1. OHLCV Data: Open, High, Low, Close, and Volume data for various time periods.

Current and Future Access Methods

Currently, we offer a pull-based S3 API for accessing flat files. This allows you to retrieve data on-demand using S3-compatible tools and libraries.

Coming Soon: Push API

We're excited to announce that we'll soon be releasing a Push API feature. With this new capability, you'll be able to:

  • Define specific data sets you need
  • Have data automatically pushed to your designated S3 bucket on a daily basis
  • Streamline your data pipeline by receiving updates without manual polling

Stay tuned for more information about this upcoming feature!

Getting Started

To start using the Flat Files:

  1. Sign up for a CoinAPI account if you haven't already.
  2. Obtain your API credentials from the dashboard.
  3. Choose an S3-compatible client or SDK.
  4. Configure your client with our endpoint and your credentials.
  5. Start retrieving historical market data!

For detailed information on authentication, billing, and API usage, please refer to the articles from the menu.

Standards and conventions

This section represents used standards and conventions across all documents and API's.


The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119.

Variables naming

All variables are named using the Snake case (or snake_case) naming convention. This means that words are separated by a single underscore _ character, no spaces are used, and letters are lowercase.

Asset codes

ISO 4217 currency code standard is used for fiat money identifications. Cryptocurrency assets are identified using codes used by the general public or adopted by the majority of exchanges.

Exchange codes

Exchange on our platform is identified by the specific exchange API and the matching engine behind it. When the example exchange has multiple separate APIs, e.g., for different products SPOT, OPTIONS, or to cover other regions, we will expose these symbols from these respective APIs on different exchange identifiers. Below are examples of BINANCE (multiple APIs, multiple regions) and DERIBIT (single API, single region). A full listing of exchanges can be queried using the Market Data REST API.

BINANCE Exchange IDsWebsiteDescription
BINANCEFTS Futures (USDT/dapi)
BINANCEFTSC Futures (Coin/fapi)
BINANCEFTSCUAT Futures Testnet (Coin/fapi)
BINANCEFTSUAT Futures Testnet (USDT/dapi)
BINANCEOPTV Options Vanilla (vapi)
BINANCEOPTVUAT Options Vanilla Testnet (vapi)
DERIBIT Exchange IDsWebsiteDescription

Numbers precision

Numbers in our platform can have a maximum of 19 digits overall, but no more than 9 decimal places. In cases when the number represents aggregate value then we allow 38 digits overall, but still no more than 9 decimal places.


For all input and output time values ISO 8601 standard is used.

Format specifierDescription
yyyyThe year is a four-digit number.
MMThe month, from 01 through 12.
ddThe day of the month, from 01 through 31.
HHThe hour, using a 24-hour clock from 00 to 23.
mmThe minute, from 00 through 59.
ssThe second, from 00 through 59.
fffThe milliseconds in a date and time value.
fffffffThe ten-millionths of a second in a date and time value.

Input time values are parsed using the following formats as far as possible:

  • yyyy-MM-ddTHH:mm:ss.fffffff
  • yyyy-MM-ddTHH:mm:ss.fff
  • yyyy-MM-ddTHH:mm:ss
  • yyyy-MM-ddTHH:mm
  • yyyy-MM-ddTHH
  • yyyy-MM-dd
  • yyyyMMddTHHmmssfffffff
  • yyyyMMddTHHmmssfff
  • yyyyMMddTHHmmss
  • yyyyMMddTHHmm
  • yyyyMMddTHH
  • yyyyMMdd

When time zone information is not supplied, we will assume the UTC zone.

Output time values are formatted using the following patterns:

  1. yyyy-MM-ddTHH:mm:ss.fffffffZ
  2. yyyy-MM-dd

All time values we provide are UTC zones. Do not assume otherwise.