# open-mmlab/mmpretrain

OpenMMLab Pre-training Toolbox and Benchmark

Repository: https://github.com/open-mmlab/mmpretrain
Canonical: https://ross.abutalabs.com/products/mmpretrain
Homepage: https://mmpretrain.readthedocs.io/en/latest/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: image-classification, resnet, mobilenet, pytorch, deep-learning, swin-transformer, beit, clip, constrastive-learning, convnext, mae, masked-image-modeling, moco, pretrained-models, self-supervised-learning, vision-transformer, multimodal
Last push: 2024-11-01T06:27:36+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2246, "days_push": 670, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3850, forks 1109 (observed 2026-08-28T04:08:24.245773+00:00)

## What it is
MMPretrain is OpenMMLab's PyTorch-based toolbox and benchmark for image classification model pre-training, covering supervised, self-supervised, and multimodal approaches. It provides a large model zoo including ResNet, Vision Transformers, CLIP, MAE, and MoCo with modular training and evaluation configs.

## Use cases
- train an image classification model on my own dataset
- benchmark pretrained vision transformers like Swin or ViT
- run self-supervised pretraining such as MAE or MoCo
- fine-tune a pretrained ResNet or ConvNeXt classifier
- extract image features with a pretrained backbone
- compare image classification models on ImageNet

## When to choose
- you need a configurable PyTorch framework for image classification or pretraining research
- you want access to a broad zoo of pretrained vision backbones with consistent APIs
- you are benchmarking supervised or self-supervised vision models

## When to avoid
- you only need simple inference without the OpenMMLab config ecosystem
- your task is detection or segmentation rather than classification or pretraining
- you prefer lightweight single-file training scripts

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, computer-vision, benchmarking
- domain: deep-learning, computer-vision, machine-learning, image-processing
- platform: python
- tags: image-classification, pretrained-models, self-supervised-learning, vision-transformer, pytorch, model-zoo, openmmlab, linux, gpu

## Member repositories
- open-mmlab/mmpretrain (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:24.245773+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 2026-08-29T18:25:58.294525+00:00, confidence not recorded.
  - readme: https://github.com/open-mmlab/mmpretrain (fetched 2026-08-28T04:08:24.245773+00:00, sha efe30bfc2c85)
  - registry_pypi: https://pypi.org/pypi/mmpretrain/json (fetched 2026-08-29T09:21:09.795757+00:00, sha 6d5a7a1cf7fd)
- Data as of 2026-08-30T08:39:29.467469+00:00.