# PaddlePaddle/Paddle

PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）

Repository: https://github.com/PaddlePaddle/Paddle
Canonical: https://ross.abutalabs.com/products/paddle
Homepage: http://www.paddlepaddle.org/
Language: C++
License: Apache-2.0
License Family: permissive
Topics: paddlepaddle, deep-learning, scalability, machine-learning, neural-network, python, efficiency, distributed-training
Last push: 2026-08-26T21:19:33+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 56, longevity 100
- inputs: {"age_days": 3670, "days_push": 7, "days_rel": 214, "gap_med": 54, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 24062, forks 6018 (observed 2026-08-28T04:11:35.537290+00:00)

## What it is
PaddlePaddle is an industrial-grade deep learning framework written in C++ with Python APIs, supporting high-performance single-machine and distributed training plus cross-platform deployment. It offers unified dynamic/static graphs, automatic parallelism, integrated large-model training and inference, and high-order differentiation for scientific computing.

## Use cases
- train deep learning models on single machine or distributed cluster
- train and fine-tune large language models with automatic parallelism
- deploy trained models across platforms for inference
- build industrial AI applications in manufacturing or agriculture
- run high-order differentiation for scientific computing
- develop neural networks with unified dynamic and static graphs

## When to choose
- you need scalable distributed training with minimal partitioning annotations
- you want one framework covering training through deployment for large models
- you work in the Chinese AI ecosystem or need Paddle ecosystem models and toolkits
- you need industrial-strength deep learning with strong production adoption

## When to avoid
- your team and tooling are standardized on PyTorch or TensorFlow
- you need the broadest third-party model and library compatibility
- you only need lightweight experimentation without distributed training

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, deep-learning, llm-training, llm-inference, gpu-computing
- domain: deep-learning, machine-learning, large-language-models, gpu-computing
- platform: windows, python, cpp
- tags: paddlepaddle, distributed-training, neural-networks, automatic-parallelism, scientific-computing, chinese-ai-ecosystem, linux, macos, gpu, docker

## Member repositories
- PaddlePaddle/Paddle (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:35.537290+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 2026-08-29T16:56:17.616776+00:00, confidence not recorded.
  - readme: https://github.com/PaddlePaddle/Paddle (fetched 2026-08-28T04:11:35.537290+00:00, sha 901644c98143)
- Data as of 2026-08-30T08:39:29.467469+00:00.