Pytubefix

Python3 Library for Downloading YouTube Videos

---

Installation

pip install pytubefix

---

Quickstart

Download MP4 Video in Highest Resolution:

from pytubefix import YouTube
from pytubefix.cli import on_progress

url = "url"

yt = YouTube(url, on_progress_callback=on_progress)
print(yt.title)

ys = yt.streams.get_highest_resolution()
ys.download()

Download Audio-Only (.m4a):

from pytubefix import YouTube
from pytubefix.cli import on_progress

url = "url"

yt = YouTube(url, on_progress_callback=on_progress)
print(yt.title)

ys = yt.streams.get_audio_only()
ys.download()

Download a Complete Playlist:

from pytubefix import Playlist
from pytubefix.cli import on_progress

url = "url"

pl = Playlist(url)
for video in pl.videos:
    ys = video.streams.get_audio_only()
    ys.download()

Use OAuth Authentication:

from pytubefix import YouTube
from pytubefix.cli import on_progress

url = "url"

yt = YouTube(url, use_oauth=True, allow_oauth_cache=True, on_progress_callback=on_progress)
ys = yt.streams.get_highest_resolution()
ys.download()  # Authenticate once for subsequent downloads

Specify Output Directory for Downloads:

from pytubefix import YouTube
from pytubefix.cli import on_progress

url = "url"

yt = YouTube(url, on_progress_callback=on_progress)
ys = yt.streams.get_highest_resolution()
ys.download(output_path="path/to/directory")

---

Working with Subtitles/Caption Tracks

View Available Subtitles:

from pytubefix import YouTube

yt = YouTube('http://youtube.com/watch?v=2lAe1cqCOXo')
print(yt.captions)

Print Subtitle Tracks:

from pytubefix import YouTube

yt = YouTube('http://youtube.com/watch?v=2lAe1cqCOXo')
caption = yt.captions['a.en']
print(caption.generate_srt_captions())

Save Subtitles to a Text File:

from pytubefix import YouTube

yt = YouTube('http://youtube.com/watch?v=2lAe1cqCOXo')
caption = yt.captions['a.en']
caption.save_captions("captions.txt")

---

Using Channels

Get Channel Name:

from pytubefix import Channel

c = Channel("https://www.youtube.com/@ProgrammingKnowledge/featured")
print(f'Channel name: {c.channel_name}')

Download All Videos from a Channel:

from pytubefix import Channel

c = Channel("https://www.youtube.com/@ProgrammingKnowledge")
print(f'Downloading videos by: {c.channel_name}')

for video in c.videos:
    video.streams.get_highest_resolution().download()

---

Search for Videos

Basic Search:

from pytubefix import Search

results = Search('GitHub Issue Best Practices')
for video in results.videos:
    print(f'Title: {video.title}')
    print(f'URL: {video.watch_url}')
    print(f'Duration: {video.length} sec')
    print('---')

Use Filters:

from pytubefix.contrib.search import Search, Filter

filters = (
    Filter.create()
        .upload_date(Filter.UploadDate.TODAY)
        .type(Filter.Type.VIDEO)
        .duration(Filter.Duration.UNDER_4_MINUTES)
        .feature([Filter.Features.CREATIVE_COMMONS, Filter.Features._4K])
        .sort_by(Filter.SortBy.UPLOAD_DATE)
     )

s = Search('music', filters=filters)
for video in s.videos:
    print(video.watch_url)

AsyncYouTube — Advanced Guide with Complete Examples

AsyncYouTube is a fully asynchronous Python interface built on PyTubeFix, intended for developers who require complete control over YouTube video data. It provides access to video streams, metadata, chapters, key moments, and more — all without blocking your event loop.

---

Quick Start Example

A full program demonstrating basic usage:

import asyncio
from pytubefix import AsyncYouTube

URL = "YOUR_VIDEO_URL"

async def main():
    # Initialize AsyncYouTube with OAuth to handle age-restricted content
    yt = AsyncYouTube(URL, use_oauth=True, allow_oauth_cache=True)
    
    # Fetch all available streams asynchronously
    streams = await yt.streams()
    print("Available Streams:")
    for stream in streams:
        print(stream)

if __name__ == '__main__':
    asyncio.run(main())

---

Download a Specific Stream

Complete example showing download with progress and completion callbacks:

import asyncio
from pytubefix import AsyncYouTube

URL = "YOUR_VIDEO_URL"

async def main():
    def on_progress(stream, chunk, bytes_remaining):
        total = stream.filesize
        percent = (1 - bytes_remaining / total) * 100
        print(f"\rProgress: {percent:.2f}%", end="")

    def on_complete(stream, file_path):
        print(f"\n√ Done downloading: {file_path}")

    yt = AsyncYouTube(URL, use_oauth=True, allow_oauth_cache=True)

    yt.register_on_progress_callback(on_progress)
    yt.register_on_complete_callback(on_complete)

    stream = await yt.get_stream_by_itag(18) # 360p MP4 progressive stream

    print(f"Downloading: {await yt.title()}")

    stream.download(filename="my_video.mp4") # Blocking call by design

if __name__ == '__main__':
    asyncio.run(main())

Note: Always use callbacks to track progress; download() is synchronous.

---

Fetch Video Metadata

import asyncio
from pytubefix import AsyncYouTube

URL = "YOUR_VIDEO_URL"

async def main():
    yt = AsyncYouTube(URL, use_oauth=True, allow_oauth_cache=True)

    title = await yt.title()
    views = await yt.views()
    likes = await yt.likes()
    author = await yt.author()
    thumbnail = await yt.thumbnail_url()

    print(f"Title: {title}")
    print(f"Views: {views}")
    print(f"Likes: {likes}")
    print(f"Author: {author}")
    print(f"Thumbnail URL: {thumbnail}")

if __name__ == '__main__':
    asyncio.run(main())

---

Retrieve Chapters and Key Moments

import asyncio
from pytubefix import AsyncYouTube

URL = "YOUR_VIDEO_URL"

async def main():
    yt = AsyncYouTube(URL, use_oauth=True, allow_oauth_cache=True)

    chapters = await yt.chapters()
    key_moments = await yt.key_moments()

    print("Chapters:", chapters)
    print("Key Moments:", key_moments)

if __name__ == '__main__':
    asyncio.run(main())

---

Create AsyncYouTube from Video ID

import asyncio
from pytubefix import AsyncYouTube

VIDEO_ID = "YOUR_VIDEO_ID"

async def main():
    yt = AsyncYouTube.from_id(VIDEO_ID, use_oauth=True, allow_oauth_cache=True)
    streams = await yt.streams()
    print("Streams fetched from Video ID:")
    for s in streams:
        print(s)

if __name__ == '__main__':
    asyncio.run(main())

---

Best Practices

• Always await asynchronous methods: streams(), title(), views(), likes(), chapters(), key_moments().
• Use use_oauth=True to handle age-restricted content; cache tokens to minimize repeated logins.
• Wrap network calls in try/except to handle errors gracefully.
• Combine callbacks with asyncio for efficient non-blocking downloads.
• Maintain consistent program structure with main() and asyncio.run() for readability and maintainability.

---