# apple/corenet

CoreNet: A library for training deep neural networks

Repository: https://github.com/apple/corenet
Canonical: https://ross.abutalabs.com/products/corenet
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Last push: 2025-10-09T07:33:23+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 46, release rhythm 35, longevity 61
- inputs: {"age_days": 867, "days_push": 328, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7007, forks 540 (observed 2026-08-28T04:09:53.241898+00:00)

## What it is
CoreNet is Apple's deep neural network training toolkit for training standard and novel small and large-scale models, including foundation models like CLIP and LLMs, as well as vision tasks such as object classification, detection, and segmentation. It includes training and evaluation recipes and pre-trained models from Apple's published research.

## Use cases
- train large language models from scratch
- train CLIP-style vision-language foundation models
- train image classification and object detection models
- train lightweight mobile-friendly vision transformers
- reproduce Apple research models like OpenELM and FastViT
- run semantic segmentation model training

## When to choose
- you need a research-grade PyTorch training toolkit for vision and language models
- you want to reproduce or extend Apple's published model architectures
- you need training recipes and pre-trained weights for efficient small models
- you are training foundation models like LLMs or CLIP

## When to avoid
- you only need model inference rather than training
- you want a high-level framework with simple fine-tuning APIs like Hugging Face Trainer
- you need a production serving or deployment solution
- you work outside PyTorch or need non-Python tooling

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, llm-training, image-processing
- domain: deep-learning, machine-learning, computer-vision, large-language-models
- platform: python
- tags: pytorch, neural-network-training, foundation-models, computer-vision, apple, training-recipes, vision-transformers, research, linux, macos, gpu

## Member repositories
- apple/corenet (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:53.241898+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 2026-08-29T17:40:44.348659+00:00, confidence not recorded.
  - readme: https://github.com/apple/corenet (fetched 2026-08-28T04:09:53.241898+00:00, sha f4c8aab59583)
- Data as of 2026-08-30T08:39:29.467469+00:00.