# FedML-AI/FedML

FEDML - The unified and scalable ML library for large-scale distributed training, model serving, and federated learning. FEDML Launch, a cross-cloud scheduler, further enables running any AI jobs on any GPU cloud or on-premise cluster. Built on this library, TensorOpera AI (https://TensorOpera.ai) is your generative AI platform at scale.

Repository: https://github.com/FedML-AI/FedML
Canonical: https://ross.abutalabs.com/products/fedml
Homepage: https://TensorOpera.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: federated-learning, deep-learning, distributed-training, edge-ai, machine-learning, on-device-training, inference-engine, mlops, model-deployment, model-serving, ai-agent
Last push: 2025-10-28T12:44:12+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 49, release rhythm 8, longevity 100
- inputs: {"age_days": 2234, "days_push": 309, "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 4062, forks 765 (observed 2026-08-28T04:08:34.033862+00:00)

## What it is
FedML (TensorOpera) is a unified Python library for large-scale distributed training, model serving, and federated learning across GPU clouds, on-premise clusters, edge servers, and smartphones. It includes a cross-cloud scheduler (Launch) that pairs AI jobs with economical GPU resources and auto-provisions them.

## Use cases
- run federated learning across edge devices and clouds
- train large models on distributed multi-cloud GPUs
- deploy and serve models with low latency
- schedule AI jobs on the cheapest available GPU cloud
- train models on smartphones and IoT devices
- manage on-premise GPU clusters for AI workloads

## When to choose
- you need federated or cross-silo learning across heterogeneous devices
- you want one library spanning training, deployment, and multi-cloud scheduling
- you need on-device or edge training for mobile/IoT scenarios

## When to avoid
- you only need simple single-machine model training
- you want a lightweight inference-only serving tool
- you prefer a fully managed service without an open-source library

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-training, llm-inference, deployment, gpu-computing, agent-framework
- domain: machine-learning, deep-learning, large-language-models, artificial-intelligence, microservices, gpu-computing, developer-tools
- platform: python, cross-platform, cloud
- tags: federated-learning, distributed-training, mlops, edge-ai, model-deployment, cross-cloud-scheduler, on-device-training, model-serving, docker, gpu

## Member repositories
- FedML-AI/FedML (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:34.033862+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 2026-08-29T18:23:34.711335+00:00, confidence not recorded.
  - readme: https://github.com/FedML-AI/FedML (fetched 2026-08-28T04:08:34.033862+00:00, sha fb8c74cf76d6)
  - homepage: https://TensorOpera.ai (fetched 2026-08-29T09:15:46.011282+00:00, sha 365f54b1aaa4)
  - registry_pypi: https://pypi.org/pypi/fedml/json (fetched 2026-08-29T09:15:46.018683+00:00, sha b42075650907)
- Data as of 2026-08-30T08:39:29.467469+00:00.