AILab-CVC/YOLO-World

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

6529 stars · 612 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

YOLO-World is a real-time open-vocabulary object detection model and Python toolkit from Tencent AI Lab and HUST, published at CVPR 2024. It detects arbitrary objects described in text without custom training, using a fast CNN-based YOLO architecture that is significantly faster and smaller than Transformer-based open-vocabulary detectors.

Use cases

• detect objects in images by text prompt without training
• run real-time zero-shot object detection on edge devices
• auto-label image datasets for training custom detectors
• detect custom object classes in video streams in real time
• build open-vocabulary detection pipelines in Python
• segment detected objects with YOLO-World plus EfficientSAM

When to choose

• you need fast, real-time detection of classes not in a fixed label set
• you want zero-shot detection without annotating or training a custom model
• you need a lighter, faster alternative to Grounding DINO or GLIP
• you want to auto-label data or deploy detection to edge devices

When to avoid

• you need a permissively licensed model for commercial products (GPL-3.0)
• you need closed-set detection with maximum accuracy and can train a specialized model
• you need text generation or multimodal reasoning beyond detection
• you require a maintained stable API, as the project is still under active development

Facets

library · maturity active

computer-vision machine-learning image-processing deep-learning computer-vision machine-learning deep-learning artificial-intelligence python cross-platform object-detection open-vocabulary-detection zero-shot yolo real-time-inference cvpr-2024 vision-language gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="AILab-CVC/YOLO-World")