Baike.dev
All toolsOnline toolsAI codingTrendingOpen sourceNewsSubmit
Log in
< Back to tools
F

free-proxy

> 编程语言
Open source

Free proxy scraper written in python. It is pypi library - free to use.

396 stars0 likes0 views
WebsiteGitHub

About

Free proxy scraper written in python. It is pypi library - free to use.

Free-proxy

Get free working proxies from , , and and use them in your script

The FreeProxy class scrapes proxies from , , , and and checks to make sure that it works. You can filter proxies by country, and specify an acceptable timeout. You can also randomize the list of proxies, rather than going in the order that they are scraped in.

You can use this to send requests through a custom proxy, with Selenium, or with anything else.

Returns proxy as string:

python
'http://113.160.218.14:8888'

Requirements

  • Python 3 (3.9+)
  • Requests library
  • Lxml library
  • Aiohttp library (optional, only for async usage)

Installation

python
pip install free-proxy

With async support (installs aiohttp):

python
pip install "free-proxy[async]"

Usage examples

First import Free Proxy that way:

python
from fp.fp import FreeProxy

Options

Parameter Type Example Default value
country_id list ['US', 'BR'] None
timeout float > 0 0.1 0.5
rand bool True False
anonym bool True False
elite bool True False
google bool,None False None
https bool True False
url str '' google.com
request_timeout float > 0 5 10
  • No parameters Get the first working proxy from . If no proxies are working, try again pulling from
python
proxy = FreeProxy().get()
  • country_id parameter Get the first working proxy from a specified list of countries (from ). If no proxies are working check all countries pulling from .
python
proxy = FreeProxy(country_id=['US', 'BR']).get()
  • country_id for US and GB You can set the country_id to US or GB to get a proxy from the United States or the United Kingdom respectively. Proxies will be scrapped from (US) or (GB). If there are no working proxies in the specified list check all countries
python
proxy = FreeProxy(country_id=['US']).get()
proxy = FreeProxy(country_id=['GB']).get()
  • timeout parameter Timeout is the parameter for checking if a proxy is valid. If the server does not respond in specified time, the script will mark the proxy as invalid. Default timeout=0.5. You can change it by specifying a timeout eg. timeout=1.
python
proxy = FreeProxy(timeout=1).get()
  • rand parameter Shuffles the order of the proxy list from instead of going from newest to oldest (as listed on the website). Defaults to rand=False
python
proxy = FreeProxy(rand=True).get()
  • anonym parameter Return only proxies marked as anonymous. Defaults to anonym=False
python
proxy = FreeProxy(anonym=True).get()
  • elite parameter Return only proxies marked as 'elite proxy'. Defaults to elite=False.
python
proxy = FreeProxy(elite=True).get()

Please note: elite proxies are always anonymous. If you set elite=True, you will also be eliminating any non-anonymous proxies.

  • google parameter If True it will only return proxies marked as "google". If False, it will not return proxies marked as "google". Defaults to google=None, which returns all proxies.
python
proxy = FreeProxy(google=True).get()
  • https parameter If True it will only return proxies marked as HTTPS. Defaults to https=False - i.e. HTTP proxy (for HTTP websites).

    Please note: HTTPS proxies work for both HTTP and HTTPS websites.

python
proxy = FreeProxy(https=True).get()

You can combine parameters:

python
proxy = FreeProxy(country_id=['US', 'BR'], timeout=0.3, rand=True).get()

If there are no working proxies with the provided parameters, the script will raise FreeProxyException with the message There are no working proxies at this time..

  • url parameter The url parameter allows you to set a custom URL for testing purposes. If left empty, it defaults to 'https://www.google.com'.

Using default URL:

python
proxy = FreeProxy().get() 

Using custom URL, if test on different endpoint is needed:

python
proxy = FreeProxy(url='http://httpbin.org/get').get() 
  • request_timeout parameter Number of seconds to wait when downloading the proxy list from the source websites ( etc.). If the website does not respond in that time, the script raises FreeProxyException instead of hanging indefinitely. Defaults to request_timeout=10. Not to be confused with timeout, which is used for checking individual proxies.
python
proxy = FreeProxy(request_timeout=5).get()

Async usage

AsyncFreeProxy is an asyncio variant of FreeProxy. It accepts the same options, but checks proxies concurrently (up to max_concurrent at a time, default 20) and returns the first working one, which is usually much faster than the synchronous version. Remaining checks are cancelled as soon as a winner is found.

Async support requires the async extra:

python
pip install "free-proxy[async]"

In a plain script, run it with asyncio.run:

python
import asyncio

from fp.fp import AsyncFreeProxy

proxy = asyncio.run(AsyncFreeProxy().get())

Inside your own async application (or in a Jupyter cell), just await it:

python
proxy = await AsyncFreeProxy(country_id=['US'], max_concurrent=50).get()
  • max_concurrent parameter (AsyncFreeProxy only) Maximum number of proxies checked at the same time. Defaults to max_concurrent=20. Raise it for faster scanning on a strong connection, lower it to be gentler on your network and the test URL.

CHANGELOG


[1.3.0] - 2026-09-06

  • Added AsyncFreeProxy — an asyncio variant of FreeProxy that checks proxies concurrently and returns the first working one (based on the idea from #39, thanks @NerdzzyDev)
  • Async support is an optional extra: pip install "free-proxy[async]" (installs aiohttp)
  • Added max_concurrent parameter to AsyncFreeProxy (default: 20) limiting how many proxies are checked at once
  • Raised minimum Python version to 3.9

[1.2.3] - 2026-08-24

  • Fixed the proxy check silently bypassing the proxy when the test URL scheme did not match the https setting. With default settings (HTTP proxy mapping + HTTPS test URL) requests selected no proxy at all, the peername verification then rejected every candidate and get() always failed. The proxies mapping now covers both http and https schemes. Broken since 1.2.1. (#61)
  • Added a CI workflow running the test suite on Python 3.9–3.13

[1.2.2] - 2026-07-04

  • Fixed get_proxy_list hanging indefinitely when a source website stalls — the proxy list request now has a timeout (#58)
  • Added request_timeout parameter (default: 10 seconds) to control the proxy list download timeout

[1.2.1] - 2026-06-09

  • Fixed proxy check failing due to a duplicated URL schema (http://https://...), introduced in 1.1.3. This broke https=True and custom url usage (#52, #43).

[1.2.0] - 2026-06-09

  • Updated lxml to version 6.1.1
  • Updated requests to version 2.34.2
  • Updated pip-chill to version 1.0.5
  • Pinned minimum dependency versions in install_requires (`lxml>=5.0,=2.31,
  • Added new website to get proxies from
  • Added new website to get proxies from
  • Change lxml version to 4.9.1

[1.0.6] - 2022-01-23

  • Added google parameter
  • Added https parameter

[1.0.5] - 2022-01-07

  • Added elite parameter
  • Add exception class and raise exception instead of system exit
  • Change lxml version to 4.6.5

[1.0.4] - 2021-11-13

  • Fix proxy list default length

[1.0.3] - 2021-08-18

  • Change XPatch due to SSL proxies page update
  • Change lxml version

[1.0.2] - 2020-09-03

  • Added anonym parameter

[1.0.1] - 2020-03-19

  • Fix typos in readme
  • Fix urrlib3 exception urllib3.exceptions.ProxySchemeUnknown: Not supported proxy scheme None,
  • Fix imports

[1.0.0] - 2019-02-04

  • Initial release

Disclaimer

The authors of this repository are not responsible for any consequences, damages or losses arising from the use or misuse of this repository or content. Users are solely responsible for their actions and any consequences that may follow.

License


MIT

Free Software!

Issues· 8 open

View all issuesOpen on GitHub

No open issues yet, or sync has not completed.

> Tags

Pythonfree

No comments yet. Be the first to share.

> Details

PublishedAug 1, 2026
UpdatedSep 17, 2026
Category编程语言
PricingOpen source

> Related tools

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言