microsoft/GLIP

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2607 stars · 218 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

GLIP is Microsoft's official implementation of Grounded Language-Image Pre-training, a vision-language model that unifies object detection and phrase grounding by reformulating detection as a grounding task. It achieves strong zero-shot and few-shot transfer to object-level recognition tasks like COCO and LVIS detection.

Use cases

• detect arbitrary objects in images using natural language prompts
• zero-shot object detection without training on target dataset
• ground text phrases to image regions
• fine-tune a detection model on custom classes with few examples
• evaluate open-vocabulary detection benchmarks like COCO and LVIS

When to choose

• you need open-vocabulary or text-prompted object detection
• you want a research-grade pretrained vision-language detection model
• you have few labeled examples and need strong few-shot transfer

When to avoid

• you need a lightweight production detector with minimal dependencies
• you only need standard closed-set detection with fixed classes
• you lack GPU resources for large vision-language models

Facets

library · maturity maintenance

computer-vision machine-learning image-processing deep-learning computer-vision machine-learning artificial-intelligence python object-detection grounded-language-image-pretraining zero-shot-detection vision-language open-set-detection research-code linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="microsoft/GLIP")