# facebookresearch/dino

PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

Repository: https://github.com/facebookresearch/dino
Canonical: https://ross.abutalabs.com/products/dino
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2024-07-03T16:21:59+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1960, "days_push": 791, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7611, forks 1049 (observed 2026-08-28T04:10:02.026207+00:00)

## What it is
PyTorch implementation of DINO, a self-supervised learning method for training Vision Transformers, with pretrained model weights. It is the official research code from Facebook AI Research accompanying the paper 'Emerging Properties in Self-Supervised Vision Transformers'.

## Use cases
- train vision transformers with self-supervised learning
- download pretrained DINO ViT backbones for downstream tasks
- extract image features without labels
- reproduce DINO research results
- evaluate self-supervised models with k-NN or linear probing
- convert vision transformer backbones to ONNX

## When to choose
- you need self-supervised pretraining of ViTs on your own images
- you want strong frozen visual features for classification, retrieval, or segmentation
- you are reproducing or building on the DINO paper

## When to avoid
- you need the newest features and models - use DINOv2 instead
- you want a general-purpose computer vision toolkit rather than a research codebase
- you lack GPU resources for training

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing
- domain: machine-learning, computer-vision, deep-learning
- platform: python
- tags: self-supervised-learning, vision-transformer, pytorch, pretrained-models, research, gpu

## Member repositories
- facebookresearch/dino (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:02.026207+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 2026-08-29T17:36:50.142169+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/dino (fetched 2026-08-28T04:10:02.026207+00:00, sha 8c6dba3685e3)
- Data as of 2026-08-30T08:39:29.467469+00:00.