# deepjavalibrary/djl

An Engine-Agnostic Deep Learning Framework in Java

Repository: https://github.com/deepjavalibrary/djl
Canonical: https://ross.abutalabs.com/products/djl
Homepage: https://djl.ai
Language: Java
License: Apache-2.0
License Family: permissive
Topics: deep-learning, neural-network, ai, java, mxnet, machine-learning, deep-neural-networks, ml, autograd, djl, pytorch, tensorflow, onnxruntime
Last push: 2026-08-01T16:30:42+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 49, longevity 100
- inputs: {"age_days": 2500, "days_push": 32, "days_rel": 260, "gap_med": 66, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4842, forks 757 (observed 2026-08-28T04:09:01.292843+00:00)

## What it is
Deep Java Library (DJL) is an engine-agnostic, high-level deep learning framework for Java that supports backends like PyTorch, TensorFlow, MXNet, and ONNX Runtime. It lets Java developers build, train, and deploy deep learning models with a native Java API, including a pre-trained model zoo and automatic CPU/GPU selection.

## Use cases
- run deep learning inference in a Java application
- train neural networks from Java without writing Python
- load pre-trained models from a model zoo for object detection
- switch deep learning engines without changing application code
- serve ML models inside existing JVM microservices
- use GPU acceleration for Java-based ML workloads

## When to choose
- your team is Java/JVM-based and wants to avoid a Python stack
- you need engine-agnostic model loading and inference in production Java services
- you want pre-trained models with a simple Criteria-based API

## When to avoid
- you need cutting-edge research models or rapid Python-ecosystem experimentation
- your project is non-JVM and a native Python framework fits directly
- you need fine-grained low-level control over a specific engine's internals

## Facets
- artifact type: framework
- maturity: active
- function: deep-learning, machine-learning, llm-inference, image-processing, nlp
- domain: deep-learning, machine-learning, artificial-intelligence, cross-platform
- platform: jvm, cross-platform
- tags: engine-agnostic, model-zoo, pytorch, tensorflow, mxnet, onnxruntime, autograd, inference, training, gpu

## Member repositories
- deepjavalibrary/djl (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:01.292843+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 2026-08-29T18:18:27.507854+00:00, confidence not recorded.
  - readme: https://github.com/deepjavalibrary/djl (fetched 2026-08-28T04:09:01.292843+00:00, sha b7f7ebdc0909)
  - homepage: https://djl.ai (fetched 2026-08-29T09:01:07.579803+00:00, sha dcd6371cbfff)
- Data as of 2026-08-30T08:39:29.467469+00:00.