ZeroHackz/OpenFlowFrames
30Last commit Aug 13, 2026
OpenFlowFrames DSH plugin
OpenFlowFrames is a lean, free and open-source fork of Flowframes, packaged as a DSH plugin. It provides two tools—flowframes_probe and flowframes_interpolate—for inspecting video/frame directories and running RIFE interpolation headlessly. It supports any GPU via rife-ncnn-vulkan, no CUDA required.
How to install the OpenFlowFrames DSH plugin
dsh plugin add github:ZeroHackz/OpenFlowFramesCopying does not run this command. Review the repository and version before installing the OpenFlowFrames DSH plugin.
OpenFlowFrames DSH plugin data source
OpenFlowFrames DSH plugin snapshot date: Aug 16, 2026
discovered
What the OpenFlowFrames DSH plugin can do
- Probe video or frame folder: resolution, fps, frame count, audio presence.
- Interpolate video to a higher framerate using bundled RIFE models (4.9–4.26).
- Output H.264 MP4 with audio preserved, or export interpolated frames as PNG sequence.
- Handle mixed resolutions and alpha/16-bit PNGs automatically.
- Run on CPU or any GPU (AMD, Intel, NVIDIA) via Vulkan backend.
Where the OpenFlowFrames DSH plugin fits
- Smooth 24fps footage to 60fps for slow-motion or cinematic look.
- Interpolate AI-generated frame sequences to create smooth animations.
- Batch probe all videos in a folder to list their fps and frame counts.
- Convert a 24fps clip to 120fps slow-motion footage.
- Integrate with other DSH plugins like GUI-SAM2Matting for advanced video editing.
Who the OpenFlowFrames DSH plugin is for
- Video editors who need to increase frame rate for slow-motion or smooth playback.
- AI artists generating frame sequences who want to interpolate them into fluid videos.
- Developers building automated video processing pipelines with DSH agents.
OpenFlowFrames DSH plugin limitations
- Only supports Windows platform (targeted Windows executable and bat scripts).
- Requires Python 3.10+ when running as a dsh plugin (or set OPENFLOWFRAMES_PYTHON).
- GPU acceleration depends on Vulkan compatibility; older GPUs may fall back to CPU.
- Bundled RIFE models are from 4.9 to 4.26; newer models not included.
OpenFlowFrames DSH plugin: from the repository README
Quoted from the ZeroHackz/OpenFlowFrames README, the upstream source of the OpenFlowFrames DSH plugin. Copyright remains with the original authors.
This project is a fork of the original [Flowframes by n00mkrad](https://github.com/n00mkrad/flowframes). Huge thanks to him for creating and maintaining the powerful core of this application! This fork is a **lean, fully free and open** reimagining: a modern Python GUI around the latest RIFE models — no Patreon tiers, no paid builds, no legacy codebase. ### Screenshot *Main interface: pick a video or a folder of frames, choose a model and factor, interpolate.*  ## ✨ Features * **Latest RIFE Models:** RIFE 4.9 through 4.26 bundled in the repo — no external model server, fully self-contained. * **Any GPU:** Runs on AMD, Intel, and NVIDIA via `rife-ncnn-vulkan` — no CUDA or PyTorch required. * **Modern GUI:** A clean CustomTkinter dark-mode interface — no install needed beyond Python. * **Video or Frame-Folder Input:** Interpolate a video file, or a directory of PNG/JPG/WebP frames with a custom input framerate. * **MP4 or PNG Output:** Encode to H.264 MP4 (audio preserved) or export the interpolated frames as a PNG sequence. * **Robust Frame Handling:** Mixed resolutions and alpha/16-bit PNGs (common in AI-generated fr
Read the full READMERepository license: GPL-3.0
OpenFlowFrames DSH plugin questions
How do I install OpenFlowFrames as a DSH plugin?
Run `dsh plugin add github:ZeroHackz/OpenFlowFrames` in your terminal. The plugin will be available in your DSH profile. Make sure you have Python 3.10+ installed, or set the `OPENFLOWFRAMES_PYTHON` environment variable to point to a compatible Python executable.
Does OpenFlowFrames require an NVIDIA GPU?
No, it supports any GPU (AMD, Intel, NVIDIA) via the rife-ncnn-vulkan backend. It does not require CUDA or PyTorch. If your GPU supports Vulkan, it will be used for acceleration. Otherwise, it falls back to CPU.
What model versions of RIFE are included?
The plugin bundles RIFE models from version 4.9 through 4.26. You can choose a specific model when running interpolation. Newer models are not included, but you can replace the model files in the `packages/rife-ncnn` directory if desired.
Can I interpolate a folder of frames instead of a video?
Yes, you can input a directory of PNG/JPG/WebP frames. You need to provide the original framerate via the `--fps` parameter. The plugin will treat the folder as a frame sequence and interpolate accordingly.
What output formats are supported?
You can output an H.264 MP4 video (with audio preserved from the original video) or export the interpolated frames as a PNG sequence. The output mode is selected with the `--out-mode` parameter (default is mp4).