apple/corenet

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

7008 stars · 540 forks observed · 2026-10-07

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

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

library · maturity active

deep-learning machine-learning llm-training image-processing deep-learning machine-learning computer-vision large-language-models python pytorch neural-network-training foundation-models computer-vision apple training-recipes vision-transformers research linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="apple/corenet")