smile

Statistical Machine Intelligence & Learning Engine

SMILE (Statistical Machine Intelligence & Learning Engine) is a comprehensive, high-performance machine learning and AI platform for the JVM. Built on Java 25 with idiomatic APIs for Kotlin and Scala (v4.x supports Java 21; earlier versions support Java 8), SMILE covers the entire lifecycle of machine intelligence: building models with agentic AutoML, embedding algorithms directly in your application, and serving classical ML, ONNX graphs, and LLMs on your own machines.

Vision: Intelligence Where the Work Is

A decision should not require a cloud round-trip or a fragile bridge to an external Python stack. The score, the explanation, and the model update belong on the machine that already holds the facts: inside a JVM microservice, a branch server, a payment terminal, a vehicle gateway, or an air-gapped node. SMILE makes real-time, autonomous machine intelligence native to the systems enterprises already run, from resource-constrained edge gateways to multi-GPU clusters.

What We Value

• Sovereign & Self-Hosted by Default: Your data, models, API keys, and telemetry stay on your machines and your network. Everything is built to run locally and keep working when the external link is down.
• Zero-Bloat Craftsmanship: The core engine is pure, self-contained JVM bytecode built on compact, cache-friendly data structures. You get deterministic low latency and a tiny memory footprint by default, with optional plug-and-play native acceleration (OpenBLAS/LAPACK, ONNX Runtime, LibTorch/CUDA) when hardware is available.
• Explainable & Auditable in Place: High-stakes decisions cannot be black boxes. SMILE provides both exact TreeSHAP for tree ensembles and model-agnostic KernelSHAP for any classifier or regressor, computing local feature attributions in-process alongside predictions, while AutoML workflows generate readable Java, Scala, Kotlin, or Python notebooks you can inspect line by line.
• Open Source with Enterprise Backing: Open-source licensing supports academic research and community collaboration, while commercial licensing from HALO.AI removes copyleft and source-disclosure obligations for proprietary, embedded, and mission-critical production deployments.

Key Strengths

• Comprehensive Breadth in One Engine: Hundreds of algorithms spanning classification, regression, clustering, manifold learning, time series, anomaly detection, NLP, statistical testing, linear algebra, deep learning, computer vision, and native LLM inference (MiniMax M3, Qwen, Llama, tiktoken, Radix KV cache, MTP speculative decoding).
• State-of-the-Art Speed & Footprint: Engineered with primitive-backed memory layouts, Panama FFM, and the Java Vector API (SIMD), SMILE routinely outperforms R, Python, Spark, and H2O while scaling down to a 0.2 GB GraalVM native micro-container that boots in 0.4 s and idles at ~80 MiB RAM.
• On-Device Training & Adaptation: Unlike static inference-only runtimes, SMILE trains and updates models (Random Forests, Gradient Boosting, K-Means, DBSCAN, anomaly detectors) directly on live edge or application data.
• Built for Agentic & Production Workloads: Continuous batching, prefix-preserving Radix KV caches, and speculative decoding that retains its speedup under full batch load, paired with atomic model hot-swapping, Kubernetes health probes, and Prometheus metrics.

Three Pillars of SMILE

1. Core Engine: Embed in Your Application

Call classification, regression, clustering, deep learning, NLP, and visualization directly in-process from Java, Scala, or Kotlin with no separate service to operate.

2. SMILE Serve: Unified Inference Runtime

A production-ready, Quarkus-based inference server (available as JVM, CUDA GPU, and GraalVM native containers) that brings together three complementary inference capabilities in a single process:

Capability	API prefix	Description
Classic ML	/api/v1/smile	Serialized SMILE models (.sml) — classification, regression, clustering (ClusteringModel), vector quantization (VectorQuantizerModel), and anomaly detection (AnomalyModel), with local SHAP explanations (TreeSHAP & KernelSHAP)
ONNX Runtime	/api/v1/onnx	Any model in the ONNX open format (.onnx)
LLM Chat	/api/v1/chat	OpenAI-compatible chat completions, tool calling, and SSE streaming
Models	/api/v1/models	OpenAI-compatible list and retrieve for every loaded model — chat, ONNX, and SMILE
Metrics	/q/metrics	Prometheus scrape for classic, ONNX, and LLM models

A React-based web UI is bundled and served from the same process. See serve/README.md for how to build and run the service.

3. SMILE Agent Studio: AI Agent Workspace, Autonomous Bot Hub & Polyglot IDE

A local-first desktop workspace that unites a coordinated team of 9 specialized AI agents (spanning executive chief-of-staff, product management, developer evangelism, B2B sales, data science, software architecture, Java/Python engineering, and desktop GUI automation), the always-on Agent Hub (smile hub) for scheduled and event-driven background bots with human-in-the-loop approval queues, and a polyglot IDE with interactive notebooks (Java, Scala, Kotlin, and Python/Jupyter) and code/data explorers. Studio is fully extensible: drop custom agent specs (AGENT.md), skills, MCP servers, or plugins into your workspace to add domain-specific specialists and routines with zero code changes. See studio/README.md and studio/BOT.md to get your workspace and bots running in minutes.

---

Table of Contents

1. Features
2. Module Map
3. Installation
   • Maven
   • SBT (Scala)
   • Gradle (Kotlin)
   • Native Libraries (BLAS / LAPACK)
4. Quick Start
5. SMILE Studio & Shell
6. Model Serialization
7. Visualization
8. License
9. Issues & Discussions
10. Contributing
11. Maintainers
12. Gallery

---

Features

Area	Highlights
LLM	LLaMA-3 inference, tiktoken BPE tokenizer, OpenAI-compatible REST server, SSE chat streaming
Deep Learning	LibTorch/GPU backend, EfficientNet-V2 image classification, custom layer API
Classification	SVM, Decision Trees, Random Forest, AdaBoost, Gradient Boosting, Logistic Regression, Neural Networks, RBF Networks, MaxEnt, KNN, Naïve Bayes, LDA/QDA/RDA
Regression	SVR, Gaussian Process, Regression Trees, GBDT, Random Forest, RBF, OLS, LASSO, ElasticNet, Ridge
Clustering	BIRCH, CLARANS, DBSCAN, DENCLUE, Deterministic Annealing, K-Means, X-Means, G-Means, Neural Gas, Growing Neural Gas, Hierarchical, SIB, SOM, Spectral, Min-Entropy
Manifold Learning	IsoMap, LLE, Laplacian Eigenmap, t-SNE, UMAP, PCA, Kernel PCA, Probabilistic PCA, GHA, Random Projection, ICA
Feature Engineering	Genetic Algorithm selection, Ensemble selection, TreeSHAP, KernelSHAP, SNR, Sum-Squares ratio, data transformations, formula API
NLP	Sentence / word tokenization, Bigram test, Phrase & Keyword extraction, Stemmer, POS tagging, Relevance ranking
Association Rules	FP-growth frequent itemset mining
Sequence Learning	Hidden Markov Model, Conditional Random Field
Nearest Neighbor	BK-Tree, Cover Tree, KD-Tree, SimHash, LSH
Numerical Methods	Linear algebra, numerical optimization (BFGS, L-BFGS), interpolation, wavelets, RBF, distributions, hypothesis tests
Visualization	Swing plots (scatter, line, bar, box, histogram, surface, heatmap, contour, …) and declarative Vega-Lite charts

---

Module Map

Each module has its own detailed user guide. Click the README link for the module overview, or drill into individual topic guides.

base/ — Foundation

Data structures, math, linear algebra, statistical utilities, I/O

Document	Topics
README	Module overview and dependency setup
DATA_FRAME.md	DataFrame API — creation, selection, transformation
DATA_IO.md	CSV, JSON, Parquet, Arrow, JDBC, Avro readers/writers
DATA_TRANSFORMATION.md	Scalers, encoders, imputers, feature transforms
DATASET.md	Built-in benchmark and real-world datasets
FORMULA.md	R-style formula language for model matrices
DISTRIBUTIONS.md	Probability distributions (Normal, Poisson, Beta, …)
HYPOTHESIS_TESTING.md	t-test, chi-squared, ANOVA, KS-test, …
DISTANCES.md	Euclidean, Mahalanobis, Hamming, edit distance, …
NEAREST_NEIGHBOR.md	KD-Tree, Cover Tree, BK-Tree, LSH
KERNELS.md	Gaussian, polynomial, Laplacian, and other kernel functions
RBF.md	Radial basis function networks
INTERPOLATION.md	Linear, cubic spline, bilinear, bicubic
GRAPH.md	Adjacency list/matrix graph, BFS/DFS, spanning trees
SORT.md	Quick sort, heap sort, counting sort, index sort
HASH.md	Locality-sensitive hashing, SimHash
RNG.md	Random number generators, sampling, permutations
BFGS.md	L-BFGS and BFGS numerical optimizers
ICA.md	Independent Component Analysis
TENSOR.md	N-dimensional array (CPU tensor without LibTorch)
WAVELET.md	DWT, CWT, and wavelet families
GAP.md	GAP statistic for optimal cluster count estimation
COMPRESSED_SENSING.md	Compressed sensing and basis pursuit

core/ — Machine Learning Algorithms

Classification, regression, clustering, manifold learning, and more

Document	Topics
README	Module overview
CLASSIFICATION.md	SVM, Random Forest, AdaBoost, GBDT, KNN, Naïve Bayes, LDA, …
REGRESSION.md	SVR, Gaussian Process, LASSO, Ridge, ElasticNet, GBDT, …
CLUSTERING.md	K-Means, DBSCAN, BIRCH, SOM, Spectral Clustering, …
FEATURE_ENGINEERING.md	Feature selection, PCA, ICA, projection, encoding
MANIFOLD.md	t-SNE, UMAP, IsoMap, LLE, Laplacian Eigenmap
ANOMALY_DETECTION.md	IsolationForest, one-class SVM, local outlier factor
ASSOCIATION_RULE_MINING.md	FP-growth, association rules, frequent itemsets
SEQUENCE.md	HMM (Baum-Welch, Viterbi), CRF
TIME_SERIES.md	ARIMA, box-plots, autocorrelation
REGRESSION.md	Full regression API reference
TRAINING.md	Cross-validation, bootstrap, hyper-parameter search
VALIDATION.md	Hold-out, k-fold, leave-one-out evaluation
VALIDATION_METRICS.md	Accuracy, AUC, F1, RMSE, MAE, confusion matrix
HYPER_PARAMETER_OPTIMIZATION.md	Grid search, random search, Bayesian optimization
VECTOR_QUANTIZATION.md	LVQ, Neural Gas, SOM as vector quantizers

deep/ — Deep Learning & LLMs

LibTorch-backed GPU/CPU tensor operations, neural network layers, LLaMA-3 inference, EfficientNet, ONNX Runtime & GenAI

Document	Topics
README	Core deep-learning guide (tensors, layers, loss, optimizer, model, metrics, data loading)
LLM.md	smile.llm — tokenizer, LLaMA-3, Qwen3.5, inference engine, KV cache, quantization, tensor parallelism
VISION.md	smile.vision — transforms, ImageDataset, EfficientNet-V2, ImageNet labels
ONNX.md	ONNX Runtime inference (smile.onnx) — sessions, tensors, execution providers
ONNX_GENAI.md	ONNX Runtime GenAI (smile.onnx.genai) — generate loop, chat model, multimodal

The deep/README.md covers:

• smile.deep.tensor — Tensor factory, indexing, arithmetic, AutoScope memory management, dtype/device
• smile.deep.layer — Linear, Conv2d, pooling, normalization (BN/GN/RMS), dropout, embedding, sequential blocks
• smile.deep.activation — ReLU, GELU, SiLU, Tanh, Sigmoid, Softmax, GLU, HardShrink, …
• smile.deep.Loss — MSE, cross-entropy, BCE, Huber, KL, hinge, and more
• smile.deep.Optimizer — SGD, Adam, AdamW, RMSprop
• smile.deep.Model — Abstract base class + training loop
• smile.deep.metric — Accuracy, Precision, Recall, F1Score with macro/micro/weighted averaging
• smile.deep.Dataset — Dataset interface, DataSampler, SampleBatch

The deep/LLM.md covers:

• smile.llm — Message, Role, FinishReason, ChatCompletion, ChatOptions, LanguageModel
• smile.llm.tokenizer — Tokenizer interface, Tiktoken BPE implementation (LLaMA-3 compatible)
• smile.llm.model.llama — Full LLaMA-3 stack: Llama.build(), generate(), chat(), streaming
• smile.llm.model.qwen — Qwen3.5 hybrid stack (Gated DeltaNet + gated attention), MTP, vision tower
• smile.llm.engine — Continuous-batching InferenceEngine, GenerationRequest/GenerationHandle, speculative decoding
• smile.llm.cache — Paged KvCachePool, RadixCache prefix reuse
• smile.llm.quant / smile.llm.parallel — FP8/NVFP4/Marlin quantization, tensor parallelism

The deep/VISION.md covers:

• smile.vision — VisionModel, ImageDataset, EfficientNet.V2S/M/L() pretrained models, ImageNet labels
• smile.vision.transform — Transform interface, ImageClassification pipeline, resize/crop/toTensor helpers

nlp/ — Natural Language Processing

Text normalization, tokenization, POS tagging, stemming, relevance ranking

Document	Topics
README	Module overview
TOKENIZER.md	Sentence splitter, word tokenizer, regex tokenizer
POS.md	Part-of-speech tagging (Brill tagger, HMM tagger)
STEM.md	Porter, Lancaster, Lovins stemmers; lemmatization
COLLOCATION.md	Bigram/trigram statistical tests, phrase extraction
RELEVANCE.md	TF-IDF, BM25, keyword extraction
TAXONOMY.md	WordNet integration, synsets, hypernyms

plot/ — Data Visualization

Swing-based interactive plots and declarative Vega-Lite charts

Document	Topics
README	Swing plotting API — scatter, line, bar, box, histogram, heatmap, surface, contour, wireframe
VEGA.md	Declarative smile.plot.vega (Vega-Lite) — JSON spec generation, web/Jupyter rendering

serve/ — Inference Server

Quarkus-based REST inference service with OpenAI-compatible API and SSE streaming

Document	Topics
README	Building and running the server, /chat/completions endpoint, SSE streaming, configuration

studio/ — Interactive Shell & Desktop IDE

An agentic IDE for data science using Python or SMILE

Document	Topics
README.md	Desktop Studio UX
CLI	CLI entry points (smile, smile shell, smile scala, smile serve)

scala/ — Scala API

Idiomatic Scala shim — concise wrappers, symbolic operators, Scala collections integration

Document	Topics
README	API overview, smile.classification, smile.regression, smile.clustering, smile.plot in Scala

kotlin/ — Kotlin API

Idiomatic Kotlin shim — extension functions, named parameters, builder DSLs

Document	Topics
README	API overview, extension functions, Kotlin-style builders
packages.md	Full package-by-package listing of all Kotlin extension functions

json/ — JSON Library (Scala)

Lightweight zero-dependency JSON library for Scala with a clean DSL

Document	Topics
README	Parsing, building, pattern matching, path navigation, serialization

spark/ — Apache Spark Integration

Use SMILE models inside Spark ML pipelines

Document	Topics
README	SmileTransformer, SmileClassifier, SmileRegressor; training and scoring in Spark DataFrames

---

Installation

Maven

<!-- Core ML algorithms -->
<dependency>
  <groupId>com.github.haifengl</groupId>
  <artifactId>smile-core</artifactId>
  <version>7.1.0</version>
</dependency>

<!-- Deep learning + LLMs (requires LibTorch) -->
<dependency>
  <groupId>com.github.haifengl</groupId>
  <artifactId>smile-deep</artifactId>
  <version>7.1.0</version>
</dependency>

<!-- Natural language processing -->
<dependency>
  <groupId>com.github.haifengl</groupId>
  <artifactId>smile-nlp</artifactId>
  <version>7.1.0</version>
</dependency>

<!-- Data visualization -->
<dependency>
  <groupId>com.github.haifengl</groupId>
  <artifactId>smile-plot</artifactId>
  <version>7.1.0</version>
</dependency>

SBT (Scala)

libraryDependencies += "com.github.haifengl" %% "smile-scala" % "7.1.0"

Gradle (Kotlin)

dependencies {
    implementation("com.github.haifengl:smile-kotlin:7.1.0")
}

Native Libraries (BLAS / LAPACK)

Several algorithms (manifold learning, Gaussian Process, MLP, some clustering) require BLAS and LAPACK.

Linux (Ubuntu / Debian)

sudo apt update
sudo apt install libopenblas-dev libarpack2-dev

macOS (Homebrew)

brew install arpack

Do not rely on DYLD_LIBRARY_PATH: macOS System Integrity Protection (SIP) strips DYLD_* variables when the JVM is started through the /usr/bin/java stub, so dyld never sees them. SMILE reaches libarpack through the FFM bindings, whose SymbolLookup.libraryLookup("arpack") calls dlopen on the bare name libarpack.dylib; this does not consult java.library.path, so -Djava.library.path=... and LD_LIBRARY_PATH have no effect either. SMILE therefore preloads the library by absolute path through smile.linalg.NativeLibrary, which finds it under /opt/homebrew/lib or /usr/local/lib with no copy or symlink. To point it somewhere else, set the arpack.native.path system property or the ARPACK_NATIVE_PATH environment variable.

Windows — pre-built DLLs are included in the bin/ directory of the release package. Add that directory to PATH.

GPU (CUDA) — make sure the LibTorch CUDA native libraries are on PATH (Windows) or LD_LIBRARY_PATH (Linux).

---

Quick Start

import smile.classification.RandomForest;
import smile.data.formula.Formula;
import smile.io.Read;

// Load data
var data = Read.csv("src/test/resources/iris.csv");

// Train a random forest
var forest = RandomForest.fit(Formula.lhs("species"), data);

// Predict
int label = forest.predict(data.get(0));
System.out.println("Predicted class: " + label);

For deep learning examples, see deep/README.md; for LLM and computer-vision examples, see deep/LLM.md and deep/VISION.md. For visualization examples, see plot/README.md.

---

SMILE Agent Studio & CLI

SMILE Agent Studio is the local-first AI agent workspace, 24/7 autonomous Bot Hub (smile hub), and polyglot IDE (Java, Scala, Kotlin, and Python/Jupyter). It pairs 9 built-in specialist agents (Frank Chief of Staff, Steve Product Manager, Julien Evangelist, Mike B2B Sales, Clair Data Scientist, Ada Architect, James Java Guru, Guido Pythonista, and Chuck Desktop Operator), human-in-the-loop batch draft queues, and zero-code extensibility for custom agents, skills, plugins, and bot routines with interactive notebooks and code/data editors. See studio/README.md, studio/BOT.md, and studio/CLI.md for full documentation.

Download a pre-packaged release from the releases page, then:

path/to/smile/bin/setup      # install required native dependencies
path/to/smile/bin/smile      # launch SMILE Agent Studio from your project directory

CLI entry points:

Command	Description
smile	Launch SMILE Agent Studio (desktop AI workspace, Bot Hub & polyglot IDE)
smile hub	Start the always-on Agent Hub & Bot runtime service (BOT.md)
smile shell	Java (JShell) REPL with all SMILE packages pre-imported
smile scala	Scala 3 REPL
smile train	Train a classification, regression, clustering, VQ, or anomaly model
smile predict	Run batch inference (with optional probabilities and SHAP explanations)
smile serve	Start the unified HTTP inference server (Classic ML, ONNX, and LLM chat)

To increase the JVM heap:

path/to/smile/bin/smile -J-Xmx30G

---

Model Serialization

Most SMILE models implement java.io.Serializable. You can serialize a trained model to disk and load it in a production environment or inside a Spark job:

// Save
try (var out = new ObjectOutputStream(new FileOutputStream("model.ser"))) {
    out.writeObject(forest);
}

// Load
try (var in = new ObjectInputStream(new FileInputStream("model.ser"))) {
    var loaded = (RandomForest) in.readObject();
}

---

Visualization

SMILE provides two visualization layers:

• smile.plot.swing — Swing-based interactive 2D/3D plots. See plot/README.md.
• smile.plot.vega — Declarative Vega-Lite charts for browsers and Jupyter. See plot/VEGA.md.

<dependency>
  <groupId>com.github.haifengl</groupId>
  <artifactId>smile-plot</artifactId>
  <version>7.1.0</version>
</dependency>

---

License

SMILE uses a multi-license model that keeps academic research and open-source collaboration free while providing standard, upfront commercial licenses—administered by HALO.AI LLC—for proprietary distribution and production deployments:

Component	Open Source / Evaluation ($0)	Commercial License Tier
Core Engine (base, core, nlp, deep, plot, kotlin, scala)	GPLv3 — Free for open-source projects	OEM & Embedded ($25,000 one-time) — Perpetual right to embed and ship in proprietary products free of GPLv3 disclosure, plus Year 1 Maintenance & 10 Agent Studio seats (optional 20% / $5,000 yr renewal)
SMILE Serve (serve)	AGPLv3 — Free when complying with §13 network source disclosure	Production Runtime ($12,000 / year) — Covers up to 4 production server nodes ($250/node/mo) + internal Core Engine + 5 free Agent Studio seats, with zero AGPLv3/GPLv3 disclosure
SMILE Agent Studio (studio)	BSL 1.1 — Free for personal, academic, and non-production evaluation (converts to AGPLv3 after 4 years)	Bundled Free with Runtime (5 seats), OEM (10 seats), and Enterprise (25 seats), or $95 / seat / mo ($995/seat/yr paid annually, 5-seat standalone min)
Enterprise & Edge Fleet	—	From $50,000 / year — Volume edge-node licensing (100–5,000+ gateways), site-wide OEM rights, 25 Studio seats, and optional $15k 30-day Turnkey Pilot

All commercial tiers include Critical & High CVE resolution SLAs and a 12-month capped IP indemnity. See LICENSE and Commercial Licensing & Pricing, or contact [email protected].

---

Issues & Discussions

Channel	Purpose
GitHub Discussions	Questions, ideas, show-and-tell
Stack Overflow [smile]	Technical Q&A
Issue Tracker	Bug reports and feature requests
Online Docs	Tutorials and programming guides
Java API · Scala API · Kotlin API · Clojure API	API Javadoc

---

Contributing

Please read CONTRIBUTING.md for build and test instructions.

---

Maintainers

• Haifeng Li (@haifengl)
• Karl Li (@kklioss)

---

Gallery