Vailabel Studio

What Vailabel Studio is

Vailabel Studio is a free, open-source desktop app for annotating image, video, and multimodal data on Linux, macOS, and Windows. Annotation options include bounding boxes, polygons, points, lines, circles, free-draw, and SAM smart segmentation. Its on-device copilot uses local models to detect objects, segment masks, suggest labels, and check annotation quality. Video labeling works frame by frame and includes object tracking. Projects are stored in a local SQLite database; the site says nothing is uploaded unless users choose to connect cloud storage. Optional connections include Amazon S3, Azure Blob, and Google Cloud Storage, with credentials kept in the operating system keychain. Export formats include LabelMe JSON, COCO, YOLO, YOLO-Seg, and Pascal VOC. Project types include object detection, semantic segmentation, image classification, keypoints, and mixed shapes. The copilot works with LM Studio, Ollama, and Jan, using ONNX Runtime with optional CUDA. The software is licensed under Apache-2.0. Team collaboration is planned, rather than listed as a current feature.