Skip to content

primpvstreq

MIT 3 1 504
7.1 million (month) Jun 01 2024 1.2.2(2026-04-03 07:11:15 ago)
605 15 60 NOASSERTION
Dec 28 2012 212.8 thousand (month) 25.5.0(2025-06-03 03:42:30 ago)

Primp is a Python HTTP client that impersonates real web browsers by replicating their TLS fingerprints, HTTP/2 settings, and header ordering. It is a lightweight alternative to curl-cffi for bypassing TLS and HTTP fingerprinting-based bot detection.

Key features include:

  • Browser impersonation Can impersonate Chrome, Firefox, Safari, Edge, and OkHttp clients by replicating their exact TLS fingerprints (JA3/JA4), HTTP/2 frame settings, header ordering, and other connection-level characteristics.
  • HTTP/2 support Full HTTP/2 support with configurable settings that match real browser behavior.
  • Lightweight Smaller and simpler than curl-cffi while providing similar impersonation capabilities. Built on Rust for performance.
  • Familiar API Provides a requests-like API with Session support, making it easy to adopt for developers familiar with the Python requests library.
  • Proxy support HTTP and SOCKS5 proxy support with authentication.
  • Cookie management Automatic cookie handling across requests within a session.

Primp fills a similar niche to curl-cffi and hrequests — HTTP clients designed to avoid TLS/HTTP fingerprinting — but takes a Rust-powered approach for better performance. It is particularly useful when you need to bypass bot detection that relies on connection-level fingerprinting without using a full browser.

treq is a Python library for making HTTP requests that provides a simple, convenient API for interacting with web services. It is inspired byt the popular requests library, but powered by Twisted asynchronous engine which allows promise based concurrency.

treq provides a simple, high-level API for making HTTP requests, including methods for GET, POST, PUT, DELETE, etc. It also allows for easy handling of JSON data, automatic decompression of gzipped responses, and connection pooling.

treq is a lightweight library and it's easy to use, it's a good choice for small to medium-sized projects where ease of use is more important than performance.

In web scraping treq isn't commonly used as it doesn't support HTTP2 but it's the only Twisted based HTTP client. treq is also based on callback/errback promises (like Scrapy) which can be easier to understand and maintain compared to asyncio's corountines.

Highlights


bypasstls-fingerprinthttp-fingerprinthttp2fast
uses-twistedno-http2

Example Use


```python import primp # Create a session that impersonates Chrome session = primp.Session(impersonate="chrome_131") # Make requests - TLS fingerprint matches real Chrome response = session.get("https://example.com") print(response.status_code) print(response.text) # POST with JSON data response = session.post( "https://api.example.com/data", json={"key": "value"}, ) # With proxy session = primp.Session( impersonate="firefox_133", proxy="http://user:pass@proxy.example.com:8080", ) response = session.get("https://example.com") # Different browser impersonation profiles for browser in ["chrome_131", "firefox_133", "safari_18", "edge_131"]: session = primp.Session(impersonate=browser) resp = session.get("https://tls.peet.ws/api/all") print(f"{browser}: {resp.json()['ja3_hash']}") ```
```python from twisted.internet import reactor from twisted.internet.task import react from twisted.internet.defer import ensureDeferred import treq # treq can be used with twisted's reactor with callbacks response_deferred = treq.get( "http://httpbin.org/get" ) # or POST response_deferred = treq.post( "http://httpbin.org/post", json={"key": "value"}, # JSON data={"key": "value"}, # Form Data ) # add callback or errback def handle_response(response): print(response.code) response.text().addCallback(lambda body: print(body)) def handle_error(failure): print(failure) # this callback will be called when request completes: response_deferred.addCallback(handle_response) # this errback will be called if request fails response_deferred.addErrback(handle_error) # this will be called if request completes or fails: response_deferred.addBoth(lambda _: reactor.stop()) # close twisted once finished if __name__ == '__main__': reactor.run() #Note that treq can also be used with async/await: async def main(): # content reads response data and get sends a get request: print(await treq.content(await treq.get("https://example.com/"))) if __name__ == '__main__': react(lambda reactor: ensureDeferred(main())) ``` ```

Alternatives / Similar


Was this page helpful?