← All posts

Finviz API: What Elite Offers and How to Screen in Python

The Finviz API comes with the paid Elite plan, at 10 requests every 5 seconds. What the free site and scrapers offer, and how to screen with a documented API.

Finviz offers an API to members of its paid Elite plan. A screen built in the Finviz screener can be downloaded as a CSV file or called through an API route tied to that screen, and a Finviz post dated 7 October 2026 sets the limit at 10 requests every 5 seconds, shared across all Export and API endpoints on one Elite account. The free Finviz plan does not include export, and the finviz package on PyPI reads Finviz’s web pages rather than an API. A script that needs its rules in code or applied to a past date can build the same kind of table from a documented data API such as xfinlink, where each ticker in a request counts as one request.

Does Finviz have an official API?

Yes, for Elite members. A Finviz post on its upgraded API, dated 7 October 2026, describes an API and Exports area reached from the account profile menu. It lists API access by product, “from Screener and Portfolio to News, Groups, and now Insider trades, Managers, Funds, and Calendars”. Each product follows four steps, which are configuring the tool, replacing the URL path, optionally customising the columns, and adding a personal token, with code examples for Python, Google Sheets, Node.js, C# and Excel.

The workflow starts on the website. The post puts it as building the screen in the Finviz screener and then choosing “a CSV export or an API-based workflow specifically for that screen”. The rate limit is the part a scheduled job has to respect. Per the same post, Elite allows “10 requests every 5 seconds” over “a rolling five-second sliding window”, the limit is shared across all Export and API endpoints on the account, and a request over it receives HTTP 429 with Retry-After: 5. Requests refused with a 429 do not count against the limit.

Access is tied to the subscription. The plan comparison on the Finviz Elite page lists “Export to Excel, CSV” as not included on the free plan and included on Elite. Checked on 16 November 2026, that page priced Elite at $39.50 a month billed monthly, or $299.50 a year billed annually after a promotional first-year price.

What about scraping Finviz or the unofficial Python packages?

The finviz package on PyPI takes a different route. The package, version 2.1.0 as of 16 November 2026, describes itself as an “Unofficial Python API for FinViz”, and its README says version 2.0.0 “fixes all scraping issues caused by FinViz website changes”. Its disclaimer reads “Using this library to acquire data from FinViz may be against their Terms of Service. Use it responsibly and at your own risk. This library is built for educational purposes.”

Finviz states its own position in finviz.com/robots.txt. Read on 16 November 2026, the file disallows /export, /api/v1/screener-export-csv and /screener?* for every user agent, with exceptions for a list of preset screens such as top gainers and new highs. A scraper that runs a custom screen is requesting paths the file asks automated clients to leave alone. The broader trade-off between parsing pages and calling a documented endpoint, including what breaks first, is covered in web scraping vs a financial data API.

For a personal look at today’s market, none of this matters much, and the Finviz site itself is quick to read. It starts to matter when a script runs on a schedule, feeds other code or keeps its results.

How do you build a Finviz-style screen with a documented API?

A Finviz screen combines a universe, a set of columns and a set of filters. In a data API those become three separate pieces of code. The universe comes from an index roster or a ticker list, the columns from batched requests to the prices and metrics endpoints, and the filters from pandas. A complete example with request counts per plan is in the guide on how to build a stock screener in Python with an API.

Writing the universe as code also makes it easy to date. The same call can return the universe as it stood on an earlier date, which is what a backtest of the screen needs, and two roster requests show how far a familiar index moves in a few years.

import xfinlink as xfl

xfl.set_api_key("YOUR_API_KEY")  # free at https://xfinlink.com/signup

then = xfl.index("djia", as_of="2019-12-31")     # 1 request
now = xfl.index("djia")                          # 1 request

left = then[~then["entity_id"].isin(now["entity_id"])]
joined = now[~now["entity_id"].isin(then["entity_id"])]
print(f"{len(left)} of {len(then)} members on 2019-12-31 have left the index")
print(left.sort_values("removed_date")[["ticker", "entity_name", "removed_date"]].to_string(index=False))
print()
print(f"{len(joined)} joined since:", ", ".join(sorted(joined["ticker"])))

Output:

7 of 30 members on 2019-12-31 have left the index
ticker                  entity_name removed_date
   XOM             EXXON MOBIL CORP   2020-08-28
   PFE                   PFIZER INC   2020-08-28
   UTX     UNITED TECHNOLOGIES CORP   2020-08-28
   WBA WALGREENS BOOTS ALLIANCE INC   2024-02-23
   DOW                      DOW INC   2024-11-07
  INTC                   INTEL CORP   2024-11-07
    VZ   VERIZON COMMUNICATIONS INC   2026-06-26

7 joined since: AMGN, AMZN, CRM, GOOGL, HON, NVDA, SHW

Seven of the 30 Dow members at the end of 2019 had left by November 2026, and a screen of 2019 run over today’s list would never have seen Exxon Mobil, Intel or Verizon. The comparison runs on entity_id rather than on tickers, because a member can change its ticker while staying in the index, and a different company can later take over an old ticker. Both calls count one request each on every plan, the Free plan included, and the index membership reference covers the S&P 500 and the Nasdaq-100 as well as the Dow.

Which route fits which job?

Finviz Elite fits a screen designed by eye on the Finviz site and then fed to a spreadsheet or a script on a schedule. Its API keeps the screen where it was built, and the rate limit is published.

A documented data API fits the case where the rules themselves live in code, where the universe has to be dated, or where results need to join cleanly to prices, statements and filings by a permanent identifier. xfinlink is built from SEC EDGAR public filings and market data, returns pandas DataFrames from the Python client and JSON from the REST API, and states its limits per plan on the pricing page. The Free plan covers one ticker per request and 100 requests a day for non-commercial use, which is enough to build and test a screen of the Dow 30 before choosing a paid plan for a larger universe.

FAQ

Is there a free Finviz API?
Finviz’s own pages tie export and API access to Elite. The plan comparison lists export as not included on the free plan, checked on 16 November 2026.

What is the Finviz API rate limit?
Ten requests every five seconds for Elite members, on a rolling window shared by every endpoint on the account, according to Finviz’s post of 7 October 2026. Over the limit, the API answers 429 with Retry-After: 5.

Can a Python script screen stocks without scraping?
Yes. A data API returns the columns for a list of tickers in batches, and the filtering runs in pandas. The xfinlink Python client installs with pip install -U xfinlink and takes a free key.

Built with xfinlink — free financial data API for Python.