Video2X review

Video2X is the free option in this list. It is an open source framework on GitHub that runs two modes, filtering for upscaling and frame interpolation, using Anime4K v4, Real-ESRGAN, Real-CUGAN and RIFE through ncnn and Vulkan. Version 6 is a rewrite in C and C++ with a GUI and installer for Windows, packages for Linux and a universal AppImage, plus container images for Linux and macOS. If you lack a capable GPU, the project's Google Colab notebook can borrow an Nvidia T4, L4 or A100 for free for up to 12 hours per session. The project had 22.7k stars on GitHub on our check. Requirements are strict: the GPU must support Vulkan and the precompiled binaries need a CPU with AVX2. There is no colorization or face repair. Expect to tune settings yourself.

Who it is for

Technical users who want free upscaling with open source models

Price

Free and open source; free Google Colab option.

You can use Video2X on Google Colab for free if you don't have a powerful GPU of your own.
Quoted from the vendor pricing page on November 11, 2026.

Key features

• Upscaling with Real-ESRGAN, Real-CUGAN and Anime4K v4
• Frame interpolation with RIFE
• Windows installer with GUI, Linux packages and an AppImage
• Container images for Linux and macOS
• Google Colab notebook for free cloud GPUs

Strengths and weaknesses

• Strength: Free to use, with source code on GitHub
• Strength: Supports custom MPV-compatible GLSL shaders
• Strength: Needs no extra disk space during processing beyond the output file
• Weakness: Requires a Vulkan GPU and a CPU with AVX2
• Weakness: No native macOS app; Mac users rely on containers or Colab
• Weakness: No colorization, deinterlacing or face restoration tools

Key facts

• Price: Free
• Models: Real-ESRGAN, Real-CUGAN, Anime4K v4, RIFE
• GitHub stars: 22.7k

Where it ranks

Video2X is No. 9 in our Best AI Video Upscalers verdict.