#7777th of 105 in ML frameworks and libraries
flower
Python framework for federated learning across clients with any ML library
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 7.2k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
Flower (`flwr`) is a Python framework for building federated learning and federated analytics systems, where training runs on client devices and a server aggregates the results. It works with PyTorch, TensorFlow, Hugging Face Transformers, JAX, scikit-learn, MLX, XGBoost and others. The repo ships tutorials, quickstart examples, and community baselines that reproduce published federated learning papers.
Who it is for: ML researchers and engineers training models on decentralized data
Strengths
- Framework-agnostic: documented quickstarts for PyTorch, TensorFlow, JAX, scikit-learn, MLX, XGBoost and more
- Baselines reproduce papers such as FedProx, FedAvg, FedBN, MOON and FedNova
- Examples cover Android (TFLite), iOS (CoreML), Raspberry Pi and Jetson clients
- Examples include federated fine-tuning of LLMs, Whisper and vision transformers
Weaknesses
- README gives no hardware requirements; RAM and GPU needs depend on the chosen ML framework
- Privacy and security and scaling tutorials are listed as not yet written
- Focused on federated training; not a general-purpose model server or app
- README does not describe deployment setup; Docker images are only linked via a badge
What it needs
- GPU optional
- Models: PyTorch, TensorFlow, Hugging Face Transformers, PyTorch Lightning, scikit-learn
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |