The de facto standard for professional market data and infrastructure.
Trusted by two US regulators and powering 1,855+ institutions since 2015.
US regulator
US Regulator
Global bank
Société Générale
Global bank
HSBC
University
Northwestern
University
Cambridge
University
Stanford
Asset manager
Man Group
Quant fund
Boutique quant funds
Every dataset across equities, options, and futures
Real-time feeds, historical archives, and the reference data that holds them together. Browse by asset class.
Real-Time Streaming
Equities · Live · Mercury
Live stream + delayed feeds
The same CTA/UTP feed that powers the historical archive, delivered live over multicast or TCP from Mercury. Co-located at Equinix NY4, NY2, and NY5. The data you trade on is the data you research on, without a conversion layer in between.
Since Live
Multicast, TCP, co-located delivery
Same Mercury handler as the historical archive
AAPL
Apple Inc.
MSFT
Microsoft Corp.
GOOGL
Alphabet Inc.
AMZN
Amazon.com
TSLA
Tesla Inc.
Sample datasets
CTA/UTP (Full SIP Feed)
Equities · TAQ · Bars
26 market data datasets
The official consolidated feed covering all US equity exchanges. Full trade and quote data with up to 90 quantitative fields per bar. The real SIP feed, not a derived or synthetic reconstruction.
Since 2007
Up to 90 quantitative fields per bar
Full SIP feed (CTA/UTP), not a derived or synthetic feed
AAPL
Apple Inc.
MSFT
Microsoft Corp.
GOOGL
Alphabet Inc.
AMZN
Amazon.com
TSLA
Tesla Inc.
Sample datasets
Equity Security Master
Reference · Identifiers
18 reference datasets
Battle-hardened, built from multiple sources and tested by hundreds of firms. The ASID persistent identifier follows each security through ticker changes, mergers, delistings, and corporate restructuring. Cross-referenced against FIGI, ISIN, and the other industry identifiers your pipelines already use.
Since 2007
ASID, FIGI, ISIN, plus industry identifiers
Battle-hardened, built from multiple sources
ASID
algoseek ID
FIGI
Financial Inst.
ISIN
Intl Securities
TICKER
Ticker symbol
MIC
Market Identifier
Sample datasets
Adjustment Factors
Reference · Corporate actions
Comprehensive corporate actions and adjustment factors for accurate historical analysis. Covers splits, dividends, mergers, spinoffs, and other events that affect price continuity.
Since 2007
Full corporate actions history
Essential for accurate backtesting
Splits
Stock splits
Divs
Dividends
Mergers
M&A events
Spinoffs
Spinoff events
Rights
Rights issues
Sample datasets
IPOs
Reference · New listings
2 datasets
IPO data covering new listings, initial pricing, and first-day trading activity across US equity markets.
Since 2007
New listings and initial pricing
Complete US IPO coverage
Date
IPO date
Price
Offer price
Size
Deal size
Exchange
Listing venue
Sector
Industry
Sample datasets
Index Components
Reference · Index membership
1 dataset
Historical index composition data tracking which securities belong to major indices at any point in time. Critical for avoiding look-ahead bias in index-based research.
Since 2007
Historical index membership
Point-in-time composition, no look-ahead bias
S&P 500
S&P 500 comp.
Russell
Russell indices
DJIA
Dow Jones
Nasdaq
Nasdaq 100
Sector
Sector indices
Sample datasets
Real-Time OPRA
Options · Live · Mercury
Live stream + delayed feeds
Live OPRA stream carrying the same trade and NBBO quote data as the historical archive. Delivered over multicast from Mercury with co-location at Equinix NY4, NY2, and NY5. The real-time feed and the historical archive share schema, so strategies move from backtest to production without rewrites.
Since Live
Multicast, TCP, co-located delivery
Same Mercury handler as the historical OPRA archive
SPY
SPDR S&P 500
QQQ
Invesco QQQ
AAPL
Apple Inc.
IWM
iShares Russell
VIX
CBOE Volatility
Sample datasets
OPRA
Equity options · Full feed
11 market data datasets
Complete Options Price Reporting Authority feed covering all US equity options exchanges. Trade and quote data with NBBO, top-of-book, and full quote variations.
Since 2012
Full options trade and quote data
Complete OPRA coverage across all US options exchanges
SPY
SPDR S&P 500
QQQ
Invesco QQQ
AAPL
Apple Inc.
IWM
iShares Russell
VIX
CBOE Volatility
Sample datasets
Options Security Master
Reference · Options identifiers
4 reference datasets
Well-tested options security master mapping every options contract through its lifecycle, including expiration, strike changes, and underlying corporate actions.
Since 2012
Full contract lifecycle tracking
Proprietary options identifier system
OSI
OSI symbol
Root
Root symbol
Expiry
Expiration
Strike
Strike price
Type
Put/Call
Sample datasets
Greeks and IV
Analytics · Derived
1 dataset
Pre-computed options greeks and implied volatility surfaces. Delta, gamma, theta, vega, and rho calculated using Black-Scholes-Merton across the full options universe.
Since 2012
Greeks and IV at multiple intervals
Pre-computed across the full OPRA universe
Delta
Price sensitivity
Gamma
Delta change
Theta
Time decay
Vega
Vol sensitivity
IV
Implied vol
Sample datasets
Real-Time Futures
Futures · Live · CME Group
Live stream + delayed feeds
Live CME Group feed covering CME, CBOT, NYMEX, and COMEX as an officially licensed reseller. Delivered over the same Mercury infrastructure that writes the historical archive, so the schema of a live trade matches the schema of a 2014 trade in the backtest.
Since Live
Multicast, TCP, co-located delivery
Officially licensed CME Group reseller
ES
E-mini S&P
NQ
E-mini Nasdaq
CL
Crude Oil
GC
Gold
ZB
US T-Bond
Sample datasets
CME
Equity indices · FX · Rates
6 datasets
Chicago Mercantile Exchange. Equity indices, FX, interest rates, and agricultural commodities. algoseek is an officially licensed CME data reseller.
Since 2014
6 datasets available
Officially licensed CME reseller
ES
E-mini S&P 500
NQ
E-mini Nasdaq
6E
Euro FX
GE
Eurodollar
LE
Live Cattle
Sample datasets
CBOT
Treasuries · Agriculture
6 datasets
Chicago Board of Trade. Treasury futures, agricultural commodities including corn, soybeans, and wheat.
Since 2014
Same 6 dataset types
Full CBOT universe
ZB
US T-Bond
ZN
10-Year T-Note
ZC
Corn
ZS
Soybeans
ZW
Wheat
Sample datasets
NYMEX
Energy · Crude · Gas
6 datasets
New York Mercantile Exchange. Energy futures including crude oil, natural gas, and refined products.
Since 2014
Same 6 dataset types
Full NYMEX energy complex
CL
Crude Oil
NG
Natural Gas
RB
RBOB Gasoline
HO
Heating Oil
PA
Palladium
Sample datasets
COMEX
Metals · Gold · Silver
6 datasets
Commodity Exchange. Metals futures including gold, silver, copper, and platinum.
Since 2014
Same 6 dataset types
Full COMEX metals complex
GC
Gold
SI
Silver
HG
Copper
PL
Platinum
QO
Gold Options
Sample datasets
Real-Time Future Options
Derivatives · Live · CME Group
Live stream + delayed feeds
Live options-on-futures feed across the full CME Group complex, running on the same Mercury infrastructure that writes the historical options-on-futures archive. Same schema live as historical, same co-location at NY4, NY2, NY5.
Since Live
Multicast, TCP, co-located delivery
Same Mercury handler as the historical future-options archive
ES Opts
S&P Fut Opt
CL Opts
Crude Opt
GC Opts
Gold Fut Opt
ZB Opts
T-Bond Opt
NG Opts
NatGas Opt
Sample datasets
Options on Futures
Derivatives · CME Group
4 datasets
Options on futures contracts across the full CME Group exchange complex.
Since 2014
4 datasets
One architecture, from the exchange to your application
When a feed dies mid-session, the fix on a rented stack is vendors phoning each other while your desk waits. algoseek runs everything end to end, on servers it builds, networks it owns, and a ticker plant with no third-party dependencies. What breaks behind the scenes never reaches you.
Exchanges
Raw multicast feeds
SIP (CTA/UTP) · OPRA
CME · CBOT · NYMEX · COMEX
OTC Markets · CBOE Indices · CFE
New Jersey · NY2 / NY4 / NY5
A feed
Lossless
capture
B feed
Lossless
capture
Regional failover
Chicago · CH1 / CH2
A feed
Lossless
capture
B feed
Lossless
capture
Mercury ticker plant · 3rd gen
Four-way
arbitration
4 copies in
1 consensus out
Normalization
Schema · ASID
Up to 90 fields per bar
One source
Real-time
Streaming feed
Delayed
15-minute
Historical archive
Same handler
Since 2007
By latency need
Co-location
Cross-connect
direct from Mercury
Cloud
AWS · Google · Azure
Internet
API · bulk download
Your
applications
Co-lo · Cloud
On-prem
We run the infrastructure. The data is what it produces.
algoseek was spun off from a trading fund that built its own ticker plant, co-location, and data pipeline before it ever sold a dataset. Today, the same expert engineering team operates infrastructure for funds, bulge bracket banks, fintechs, regulators, and the data vendors who build their own products on top of it.
Dedicated servers
Pre-built, ready to rack
When a team needs a couple of dedicated servers and not a project, choose a standardised small, medium, or large build. Priced and provisioned without a custom scoping cycle, hosted at Equinix NY4, NY2, NY5, or Chicago CH1, monitored around the clock.
Custom infrastructure
Built to the requirement
For enterprise deployments and complex co-location hosting, we design and build the full environment: network, feeds, compute, and remote hands. This is the work we do for banks and regulators, where the requirement does not fit a catalog and the margin for error is zero.
Data engineering and delivery
Custom fields, third-party source integration, order book aggregation, and delivery into AWS, Snowflake, Databricks, or wherever your team already works.
For data suppliers
Vendors build, host, deliver, and support their own products on algoseek infrastructure, with marketplace listing, billing, a customer console, and first-line support handled for them.
From historical research to live trading on the same data
A strategy passes through historical research, out-of-sample testing, paper trading, and live production before real capital is at risk. Most vendors cover one stage well. algoseek covers all four on the same pipeline, so the logic you validated in research is the logic that runs with real money.
Stage 01
Historical research
Backtest on 20+ years of tick data with up to 90 quantitative fields per bar. The archive is queryable from day one. No ingestion pipeline to build, no schema decisions to make, no per-query billing to watch.
Depth
20+ years, tick-level
Fields
Up to 90 per bar
Access
Notebook, SQL, download
Stage 02
Out-of-sample testing
Walk-forward validation on held-out periods, using the same fields and the same security master your strategy will see in production. When the out-of-sample test passes, it passed on the real data, not on a normalized approximation.
Isolation
Clean date-range queries
Schema
Identical to Stage 01
Reference
Point-in-time security master
Stage 03
Paper trading
Test the strategy on live feeds before you risk real money. Paper trading runs on the same Mercury multicast you will deploy on, with the same schema as the archive you backtested against, so a pass here means it passed on the real thing. Run it from the cloud, a co-located rack, or a hybrid of both.
Feed
Mercury real-time multicast
Schema
Same bar format as historical
Latency
Co-located NY4, NY2, NY5
Stage 04
Live trading
Ship to production without rewriting a single line of data-handling code. The pipeline is the same one your team has been using since Stage 01. When latency starts to matter, move into full co-location or a hybrid deployment that keeps part of the workload in the cloud, on the same feeds, without touching your strategy code.
Uptime
No client downtime since 2015
Delivery
Co-location, hybrid, multicast, TCP
Support
Core engineering team
What stays constant
The work you did in Stage 01 still runs in Stage 04.
Identifier
ASID (persistent)
Schema
Up to 90 fields per bar
Reference data
Security master, corporate events
Source
Mercury, one handler
When a strategy moves from backtest to live, nothing about the data pipeline changes, because there is only one pipeline. Same identifier, same schema, same reference data, written by the same handler from 2007 through to the tick that arrived this morning.
Explore the data before you talk to anyone
Up to a year of production data, queryable now in Python or SQL. No credit card, no sales call. The sandbox runs on ArdaDB: over 10 petabytes of algoseek data, pre-loaded and queryable. No schema to map, no batch jobs, no datasets to ingest. You connect and the data is already there.
Cost is a fixed monthly fee, not a meter. Hit the database as hard as you like and the price does not move.
What that looks like on your invoice
ArdaDB
Subsecond
typical query
Flat
monthly fee
None
per-query cost
Typical cloud data warehouse
Seconds+
typical query
Usage
monthly fee
$
per-query cost
ArdaDB runs on fixed monthly compute with AWS egress at pass-through rates. No per-query fees, no surprise costs from analysts leaving dashboards running overnight.
Updated nightly
Ingestion, quality control, and reference-data joins run by the algoseek database team.
Standard SQL
No DSL, no proprietary query language, no retraining required.
Production-proven
Powers the RESTful API, the Python libraries, and the data sandbox behind this page.
What separates the data partner from the data vendor
Five places where ordinary vendor data quietly breaks, and what we do instead.
The first thing
Support that separates a partner from a vendor
Ask most vendors a hard question and you reach a triage queue, then a script, then a wait. Ask algoseek about an OPRA symbology edge case or a condition-code interpretation and you reach someone who built the data. When a feed question is blocking your strategy, that difference is the whole relationship.
Direct to engineering
No tier-1 layer between you and the person who built it.
From the trading side
Core team rotates out of research and development.
Edge-case fluent
OPRA symbology, condition codes, corporate actions.
One business day
Contractual SLA. Most tickets close same-day.
Four-way arbitration
Exchange feeds go down. Yours stays up.
Every feed arrives four times over, an A feed and a B feed from two independent providers, each from New Jersey and Chicago. Mercury arbitrates across all four in real time and writes the consensus, so a provider or a whole region can drop without touching what reaches you.
Source paths
Live
Down
Live
Live
100%
consensus
Your stream, uninterrupted, no matter what fails behind it.
4-way
arbitration across providers and regions
A + B
two independent providers
Since 2015
no client downtime through redundancy
Identifiers that survive every corporate event
Tickers change, companies merge, share classes split. ASID is the persistent identifier that tracks each security through all of it, cross-referenced to FIGI, ISIN, and the rest. Built from multiple sources and tested by hundreds of firms.
ASID cross-reference
2012
Lists on XNAS
FB
2021
Renamed Meta Platforms
FB
2022
Ticker change
META
ASID 1048291
one identifier, unbroken through every event
Point-in-time query, as of 2018
→ FB · XNAS
Cross-referenced FIGIISINTICKERMIC
Up to 90 quantitative fields in every bar
The derived fields a quant team would otherwise spend a year building, already computed in every bar: order flow, buy and sell pressure, retail flow separation, cross-exchange differences, VWAP variants, and spread dynamics. Industry-standard bars carry ten to fifteen.
Inside one TAQ minute bar
Query every dataset with one simple API
One RESTful API reaches every dataset: historical bars, real-time snapshots, security masters, and corporate actions. Standard HTTP, native Python and C# libraries, and the same SLA that serves two US regulators.
Authentication
API key in header, per-client scoping
Rate limits
Tiered by package, burst-tolerant
Response format
JSON, CSV, Parquet streaming
Versioning
Explicit /v1 path, no silent breakage
Who We Serve
The professional finds us. The institution follows.
Professionals. Institutions. Regulators. It starts with one quant who needs clean data, and ends with the whole team running on it.
Quant trading firms
Hedge funds, prop shops, and independent quants.
Banks and institutions
Surveillance, risk, compliance, execution analytics.
Regulators
Market reconstruction and oversight on reference-complete data.
Fintechs and data platforms
Custom feeds, white-label delivery, redistribution.
Academic researchers
Universities and research institutions, citable in published work.
Start with the data, not a sales call
Explore the data now
Notebook and SQL access to real production data. No credit card.
Talk to our team
Talk to someone who built the data. Response within one business day.