IDEA-Research/T-Rex

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2699 stars · 170 forks observed · 2026-08-28

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

T-Rex is the official Python API client for T-Rex2, a generic open-set object detection model that combines text and visual prompts to detect and count objects in images. It provides example code and API access via the DeepDataSpace cloud service rather than hosting model weights locally.

Use cases

• detect objects in images using text prompts
• count objects in images automatically
• detect objects via visual prompt examples
• label datasets with automated object detection
• open-set detection of custom categories without training
• integrate T-Rex2 detection API into a Python pipeline

When to choose

• you want state-of-the-art open-set object detection with text or visual prompts
• you need automated object counting for industrial or agricultural imagery
• you want to accelerate dataset annotation with model-assisted labeling
• you are okay calling a hosted cloud API instead of running models locally

When to avoid

• you need fully offline or self-hosted inference
• you require an open-source permissive license for the model itself
• you cannot depend on an external cloud API or API key
• you need real-time detection on edge devices

Facets

library · maturity active

computer-vision image-processing machine-learning sdk computer-vision image-processing artificial-intelligence python cloud object-detection open-set-detection visual-prompting text-prompting object-counting api-client eccv2024

Member repositories

For agents

markdown · JSON · MCP: product_card(name="IDEA-Research/T-Rex")