Suggest the newest and fastest FaceFusion information!

2025-07-20

The evolution of FaceFusion, an AI-based face-swapping video generation tool, never stops.
In this blog, the latest information, usage, and notes for the latest version (v3.3.2),
We are sending out the latest information on the latest version (v3.3.2), how to use it, and what to keep an eye out for.

Recommended for

  • Users of FaceFusion on Colab
  • Users who want to know about changes when a new version is released.
  • Users who want to know “which configuration is the most stable” on a test basis
  • Want to try out Google Drive integration, output with audio, and other applications.

🆕 Summary of changes in the latest version (v3.3.2)

The major updates in v3.3.0 on the official website are as follows

  • Introduce hyperswap models by FaceFusion Labs
  • Add more deep swapper models
  • Add the ultra_sharp_2_x4 frame enhancer model
  • Boost performance for high resolution processing
  • Run multiple NSFW models to prevent false detection
  • Use the 68-point face landmarks to create face area masking
  • Control lip movements using the --lip-syncer-weight argument
  • Enhance read and write reliability for corrupted videos
  • Introduce fallback logic for unsupported audio and video encoders
  • Fix the UDP webcam mode
  • Fix bitrate range violation for libopus

👉 Especially the installation time, which used to take about 8 minutes until V3.1, has been drastically reduced to about 40 seconds.
The performance improvement is visible when you actually try it out.

📌 Information for users of previous versions (v3.2 or earlier)

We have also provided a wealth of information on stable configurations and trouble cases regarding past versions (especially v3.2.0).
Although it was possible to mount Google Drive until now, it is not possible at this time with v3.3.
This is because the output file handling has become more sophisticated, and it is no longer possible to simply modify the output file as before. We expect this to improve in the near future,
So, in general, there is no reason to choose the older version at this time.

✅ We are constantly updating the “now” information.

FaceFusion is being developed at a rapid pace, and old information on the Internet will no longer be valid in the near future.
We are constantly reviewing our postings on this site based on information that has actually been verified to work in Colab and local environments.

📬 To make sure you don’t miss the latest posts

This blog will continue to post on the following topics

  • Comparison and evaluation of different versions
  • Examples of Colab template optimization
  • Applied Tips such as video with audio, subtitles, Drive preservation, etc.
  • Experimental recordings of FaceFusion and other tools integration

🔗 Related Links

📣 Finally.

For all FaceFusion users,
We will continue to provide ” current information” that we have actually tried and tested.
We hope you will bookmark this page and follow us on social networking sites!