Back to browse
GitHub Repository

Statistical Machine Intelligence & Learning Engine

6,387 starsJava

PyTorch on Java

by pdsminer·Jun 11, 2026·2 points·0 comments

AI Analysis

●●SolidNiche GemBig Brain

LibTorch bindings bring CUDA and MPS backends to Java with LLaMA-3 inference included.

Strengths
  • Wraps native LibTorch runtime instead of re-implementing deep learning ops in Java
  • LLaMA-3 inference and tiktoken BPE tokenizer work out of the box
  • Part of SMILE's broader 6.4k-star ML framework ecosystem
Weaknesses
  • DJL and Deeplearning4j already provide JVM deep learning solutions
  • Native library dependencies complicate deployment compared to pure-Java alternatives
Category
Target Audience

Java/JVM developers needing PyTorch capabilities without leaving their ecosystem

Similar To

Deep Java Library (DJL) · Deeplearning4j · JavaCPP Presets

Post Description

The smile-deep module provides idiomatic Java API for deep learning on the JVM while still reaching CPU, CUDA, and MPS backends by wrapping the PyTorch / LibTorch C++ runtime. It also provides tiktoken BPE tokenizer, LLaMA-3 inference, EfficientNet-V2, and an image classification pipeline out of the box.

Similar Projects

Developer Tools●●Solid

Webassembly4J Run WebAssembly from Java

Unified WASM API with auto-switching Panama FFI for Java 23 plus legacy JNI support.

Niche GemBig Brain
tegmentum
202mo ago