Runs on your machine
Tools built for real work
Install a dataset linter, download a video redaction app, or open browser tools for ONNX models, polygon zones, and YOLO26 LiteRT inference. Your files stay on your own machine.
- Browser tool
ONNX Visualizer & Model Inspector
View, analyze and inspect ONNX models with architecture diagrams
- Health, optimisation and runtime-compatibility analysis
- Visualize the neural network graph; benchmark and diff models
- Export architecture diagrams as PNG, SVG, AVIF, JPG or animation
Any modern browserNothing to installOpen → - Desktop appNew
Redact
Hide faces and objects in video, on your device
- On-device YOLO26 segmentation; nothing is uploaded
- A live webcam feed, or a saved video or image
- Box or mask, pixelate or blur, across 80 classes
macOSv1.0.0Open → - Command lineNew
CVFlow
Audit a detection dataset before it wastes a training run
- 27 checks: corrupt files, bad boxes, duplicates, leakage
- Terminal report, browser dashboard or one HTML file
- Fix YOLO labels in place; --strict gates a CI build
Windows, macOS & Linuxv0.0.3Open →
Browser tool · Live webcam
Try inference YOLO26 LiteRT web inference
Start your webcam to see live YOLO26 detections. The model loads automatically.
Browser tool · No upload
Draw a zone Polygon Zone Drawer
Draw counting zones on an image or video frame and copy pixel coordinates for YOLO and OpenCV.
Questions about these tools
- Are these tools free?
- Yes, all of them, with no sign-up, no trial period and no paid tier. CVFlow installs from PyPI with pip install cvflow, Redact is a free download for macOS, and the ONNX visualizer, Zone Drawer, and YOLO26 LiteRT demo run in your browser with nothing to install.
- Is my data uploaded to a server?
- No. CVFlow reads the dataset on your disk and serves its dashboard from 127.0.0.1, Redact runs its detection model on your Mac's own Neural Engine, and the ONNX visualizer parses your model inside the browser tab. Zone Drawer processes images and video frames in the browser; the YOLO26 LiteRT demo runs inference on webcam frames in the browser too. None of these tools uploads your files.
- Which platforms do they run on?
- CVFlow runs anywhere Python 3.9 or newer does, so Windows, macOS and Linux. Redact is macOS 14 or later, on Apple silicon and Intel. The ONNX visualizer, Zone Drawer, and YOLO26 LiteRT demo need a modern browser; the visualizer works best on desktop, since the model is held in the tab's memory.
- Are they open source?
- CVFlow is, under the MIT license, on GitHub. Redact ships as a signed macOS build rather than as source. The ONNX visualizer, Zone Drawer, and YOLO26 LiteRT demo run as part of this site rather than as separate packages.
- How do I hear about new ones?
- Each tool's page has a one-line email sign-up at the bottom. It is used for one thing: telling you when the next tool ships. No newsletter, no schedule, nothing else.