flwrlabs/flower

Health v2 · maintenance only

Adoption not part of the score

7085 stars · 1221 forks observed · 2026-08-28

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

Flower (flwr) is an open-source Python framework for building federated and collaborative AI systems, supporting any ML framework such as PyTorch, TensorFlow, JAX, and scikit-learn. It includes simulation and deployment runtimes, a hub for sharing federated apps, and an experimental agent runtime for collaborative AI applications.

Use cases

• train a model across distributed data without centralizing it
• run federated learning simulations locally
• federated fine-tuning of LLMs
• deploy federated learning on mobile devices like Android and iOS
• implement custom federated learning aggregation strategies
• add differential privacy to federated training
• build collaborative AI agents that share context across a federation
• federated analytics with pandas

When to choose

• you need privacy-preserving training across siloed or edge data
• you want a framework-agnostic federated learning library with strong research extensibility
• you need to simulate federated setups before real deployment
• you target heterogeneous clients including mobile and embedded devices

When to avoid

• you only need standard centralized training on a single dataset
• you need a fully turnkey production MLOps platform rather than a framework
• you depend on the experimental Flower Agent features in a stability-critical system

Facets

framework · maturity active

machine-learning deep-learning llm-training agent-framework rpc sdk machine-learning artificial-intelligence large-language-models microservices privacy python cross-platform cpp federated-learning federated-analytics pytorch tensorflow scikit-learn grpc differential-privacy simulation collaborative-ai ai-agents android ios docker kubernetes gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="flwrlabs/flower")