# Eclipse Deeplearning4J

Suite of tools for deploying and training deep learning models using the JVM. Highlights include model import for keras, tensorflow, and onnx/pytorch, a modular and tiny c++ library for running math code and a java based math library on top of the core c++ library. Also includes samediff: a pytorch/tensorflow like library for running deep learn...

Repository: https://github.com/deeplearning4j/deeplearning4j
Canonical: https://ross.abutalabs.com/products/eclipse-deeplearning4j
Homepage: http://deeplearning4j.konduit.ai
Language: Java
License: Apache-2.0
License Family: permissive
Topics: java, gpu, deeplearning, neural-nets, deeplearning4j, dl4j, hadoop, spark, intellij, artificial-intelligence, python, scala, clojure, linear-algebra, matrix-library
Last push: 2026-08-25T01:22:59+00:00
Link (homepage): http://deeplearning4j.konduit.ai
Link (site_page): https://deeplearning4j.konduit.ai/nd4j/quickstart
Link (site_page): https://deeplearning4j.konduit.ai/deeplearning4j/quickstart
Link (site_page): https://deeplearning4j.konduit.ai/python4j/getting-started
Link (site_page): https://deeplearning4j.konduit.ai/v/en-1.0.0-m1.1

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 4663, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14246, forks 3826 (observed 2026-08-28T04:11:06.615686+00:00)

## What it is
Eclipse Deeplearning4J is an open-source deep learning framework and ecosystem for the JVM, including the ND4J linear algebra library, the SameDiff automatic differentiation framework, DataVec ETL, and model import for Keras, TensorFlow, ONNX, and PyTorch. It supports CPU (AVX) and CUDA GPU acceleration, distributed training on Apache Spark, and Python interop via Python4J.

## Use cases
- train neural networks in java
- import keras or tensorflow models into a jvm application
- run deep learning inference on the jvm without python
- distributed deep learning training on apache spark
- numpy-like linear algebra library for java
- deploy deep learning models in jvm microservices or on android
- retrain pytorch or onnx models from java

## When to choose
- your production stack is JVM-based (Java, Scala, Kotlin, Clojure) and you want to train or run deep learning models without a Python runtime
- you need to import and retrain models built in Python frameworks and deploy them in JVM environments, Spark clusters, mobile, or IoT
- you need a high-performance N-dimensional array and linear algebra library for the JVM with CPU and CUDA GPU backends

## When to avoid
- your workflow is Python-first and you have no JVM requirements - PyTorch or TensorFlow are more natural choices
- you need the very latest research models or eager execution features that mature Python frameworks get first
- you want a large community and ecosystem of pretrained models and tutorials comparable to mainstream frameworks

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, etl, gpu-computing, llm-training, math
- domain: deep-learning, machine-learning, gpu-computing, data-science
- platform: jvm, windows
- tags: nd4j, samediff, datavec, keras-import, onnx, tensorflow-import, spark, hadoop, linear-algebra, model-import, python4j, scala, kotlin, clojure, natural-language-processing, linux, macos, gpu, docker, android

## Member repositories
- deeplearning4j/deeplearning4j (main) score 77
- deeplearning4j/deeplearning4j-examples (examples) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:06.615686+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 2026-08-29T17:12:42.041367+00:00, confidence not recorded.
  - readme: https://github.com/deeplearning4j/deeplearning4j (fetched 2026-08-28T04:11:06.615686+00:00, sha 94a7ded9c575)
  - homepage: http://deeplearning4j.konduit.ai (fetched 2026-08-29T08:06:23.305615+00:00, sha d3d434658ba3)
  - site_page: https://deeplearning4j.konduit.ai/nd4j/quickstart (fetched 2026-08-29T08:06:23.330131+00:00, sha 0762afc2aeb9)
  - site_page: https://deeplearning4j.konduit.ai/deeplearning4j/quickstart (fetched 2026-08-29T08:06:23.358834+00:00, sha 704448effbea)
  - site_page: https://deeplearning4j.konduit.ai/python4j/getting-started (fetched 2026-08-29T08:06:23.361646+00:00, sha 370e9c87de7d)
  - site_page: https://deeplearning4j.konduit.ai/v/en-1.0.0-m1.1 (fetched 2026-08-29T08:06:23.363853+00:00, sha f721903197be)
  - site_page: https://deeplearning4j.konduit.ai/v/en-1.0.0-beta7 (fetched 2026-08-29T08:06:23.369419+00:00, sha 2cd6ec17436d)
  - site_page: https://deeplearning4j.konduit.ai/v/en-1.0.0-beta6 (fetched 2026-08-29T08:06:23.371438+00:00, sha 6ff318fd9765)
- Data as of 2026-08-30T08:39:29.467469+00:00.