#6060th of 126 in ML frameworks and libraries
smile
JVM machine learning library with classic ML, deep learning and LLM inference
- Stars
- 6.4k
- License
- custom license
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Java
6,327 on 4 Apr 20266,428 on 31 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
SMILE is a machine learning library for the JVM with Java, Scala and Kotlin APIs. It covers classification, regression, clustering, manifold learning, NLP and plotting, plus a LibTorch-backed deep module with LLaMA-3 and Qwen3.5 inference. It also ships a Quarkus inference server (SMILE models, ONNX, OpenAI-compatible chat) and an agentic desktop IDE, SMILE Studio.
Who it is for: JVM developers building ML into Java, Scala or Kotlin applications
Strengths
- Broad classic ML coverage: SVM, random forest, GBDT, t-SNE, UMAP, HMM, CRF, FP-growth
- Serve module exposes SMILE, ONNX and OpenAI-compatible chat endpoints with a bundled web UI
- Idiomatic Scala and Kotlin APIs plus a Spark integration module
- Per-module docs and Maven Central releases (7.1.0)
Weaknesses
- Multi-product license with commercial terms; not a single standard open-source license
- v5+ requires Java 25, which limits use on older JVM stacks
- Several algorithms need native BLAS/LAPACK and ARPACK installed separately
- Deep and LLM features need LibTorch; LLM support lists only LLaMA-3 and Qwen3.5
What it needs
- GPU optional
- Needs Java 25, OpenBLAS, ARPACK, LibTorch (deep module)
- Models: SMILE .sml, ONNX, LLaMA-3, Qwen3.5, EfficientNet-V2
Quick answers
Is smile open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Does smile need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |