TNTwise Archives REAL Video Enhancer, Freezing Open-Source AI Video Tool
TNTwise flipped the popular open-source video interpolation and upscaling GUI to read-only at version 2.4.1, freezing a Flowframes alternative.
PRO- TNTwise archived REAL Video Enhancer at version 2.4.1, making the repo read-only.
- The GUI combines RIFE interpolation, Real-ESRGAN/SPAN/AnimeSR upscaling, plus denoise and decompression models.
- Three backends supported: TensorRT for RTX, PyTorch for CUDA/ROCm, and NCNN for any Vulkan 1.3 GPU.
- Positioned as a cross-platform alternative to abandoned tools like Flowframes and enhancr.
- Existing builds remain available via Flathub, Steam, and Git clone.
- Minimum 4 GB VRAM with NCNN; 8 GB VRAM recommended for TensorRT engine compilation.
The maintainer of REAL Video Enhancer, TNTwise, has archived the repository, freezing the project as read-only at version 2.4.1. The tool had grown into one of the more capable open-source GUIs for AI video processing on desktop, and its archival is worth a look if you rely on it or were planning to.
REAL Video Enhancer is a redesigned build of the original Rife ESRGAN App for Linux, offering frame interpolation and upscaling on Windows, Linux, and macOS as an alternative to aging options like Flowframes or enhancr. Think of it as a desktop wrapper that stitches together the messy pipeline of frame extraction, neural inference, and re-encoding into a single GUI backed by FFmpeg.
What the tool actually does
The project rolls four AI video tasks into one interface: interpolation (boosting frame rate), upscaling (boosting resolution), decompression (undoing codec artifacts), and denoising. Typical jobs include pushing 24 fps footage to 48 fps or upscaling 1920-wide video to 3840, which covers the two most common enhancement passes people throw at grainy source material.
Instead of shipping a single proprietary model, it wires up a menu of well-known open weights and lets you pick per job:
This story is for Pro members
You've reached the end of the free preview. Upgrade to AlphaSignal Pro to read the full article - and everything else behind the paywall.
Comments
