# sunsmarterjie/yolov12

[NeurIPS 2025] YOLOv12: Attention-Centric Real-Time Object Detectors

Repository: https://github.com/sunsmarterjie/yolov12
Canonical: https://ross.abutalabs.com/products/yolov12
Language: Python
License: AGPL-3.0
License Family: copyleft
Last push: 2026-05-14T20:07:15+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 82, release rhythm 40, longevity 40
- inputs: {"age_days": 573, "days_push": 111, "days_rel": 428, "gap_med": 26, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2947, forks 431 (observed 2026-08-28T04:07:31.792626+00:00)

## What it is
YOLOv12 is a PyTorch implementation of attention-centric real-time object detectors, published at NeurIPS 2025. It provides detection models plus classification and instance segmentation variants, with a faster YOLOv12-turbo version.

## Use cases
- train a custom object detection model
- real-time object detection with attention
- instance segmentation on images
- image classification with yolo models
- deploy a fast object detector
- detect objects in images with python

## When to choose
- you need state-of-the-art real-time detection with attention-based architecture
- you want the official efficient YOLOv12 implementation rather than the ultralytics port
- you need detection, segmentation, and classification from one model family

## When to avoid
- you need a permissively licensed model for closed-source products (AGPL-3.0)
- you need a battle-tested production framework with broad ecosystem support
- you only need simple pretrained inference without training

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, machine-learning, deep-learning, image-processing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python, cross-platform
- tags: object-detection, yolo, attention-mechanism, real-time-inference, instance-segmentation, image-classification, research-code, pytorch, gpu

## Member repositories
- sunsmarterjie/yolov12 (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:31.792626+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:32:47.085925+00:00, confidence not recorded.
  - readme: https://github.com/sunsmarterjie/yolov12 (fetched 2026-08-28T04:07:31.792626+00:00, sha 67b984504704)
- Data as of 2026-08-30T08:39:29.467469+00:00.