# PaddlePaddle/PaddleX

All-in-One Development Tool based on PaddlePaddle

Repository: https://github.com/PaddlePaddle/PaddleX
Canonical: https://ross.abutalabs.com/products/paddlex
Homepage: https://paddlepaddle.github.io/PaddleX/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: classification, segmentation, deployment, ocr, time-series, pp-chatocr, ai-pipelines, object-detection, formula-recognition, layout-detection, pdf2markdown, speech-recognition
Last push: 2026-06-25T04:47:54+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 90, longevity 100
- inputs: {"age_days": 2366, "days_push": 69, "days_rel": 69, "gap_med": 12, "n_releases_24m": 38}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6251, forks 1217 (observed 2026-08-28T04:09:40.994409+00:00)

## What it is
PaddleX is a low-code, all-in-one AI development tool built on the PaddlePaddle framework, bundling 200+ pretrained models into 33 production-ready pipelines covering OCR, object detection, image classification, segmentation, time-series forecasting, and document understanding. It supports the full workflow from training to high-performance inference and service-oriented deployment across diverse hardware including NVIDIA GPUs, Ascend, Cambricon, and Kunlun chips.

## Use cases
- run ocr on scanned documents
- convert pdf to markdown
- train an object detection model
- deploy a pretrained model as an api service
- forecast time series with pretrained models
- extract tables and layout from documents
- recognize formulas in images
- run speech recognition on audio

## When to choose
- you want out-of-the-box pretrained models with a simple Python API
- you need end-to-end pipelines from training to deployment on PaddlePaddle
- you must support Chinese domestic hardware like Ascend or Kunlun
- you need document intelligence features like OCR, layout analysis, and PDF parsing

## When to avoid
- you are committed to PyTorch or TensorFlow ecosystems
- you need to train custom architectures from scratch rather than fine-tune
- you want a lightweight single-purpose library instead of a full toolkit

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, ocr, computer-vision, speech-recognition, llm-inference, deployment, data-science, image-processing, pdf
- domain: machine-learning, deep-learning, computer-vision, developer-tools, artificial-intelligence
- platform: windows, python, cross-platform
- tags: low-code, pretrained-models, paddlepaddle, model-pipelines, inference, object-detection, time-series, document-parsing, model-training, natural-language-processing, linux, macos, gpu, docker

## Member repositories
- PaddlePaddle/PaddleX (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:40.994409+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 2026-08-29T17:47:08.726514+00:00, confidence not recorded.
  - readme: https://github.com/PaddlePaddle/PaddleX (fetched 2026-08-28T04:09:40.994409+00:00, sha 24fc43e52fe2)
  - homepage: https://paddlepaddle.github.io/PaddleX/ (fetched 2026-08-29T08:43:18.948415+00:00, sha 36c6c3c2e4f9)
- Data as of 2026-08-30T08:39:29.467469+00:00.