OPEN SOURCE

Built in the Open

Kompile is open source under the Apache 2.0 license. The entire stack — CLI, application layer, model server, ML runtime — is available for you to inspect, build, and contribute to.

View on GitHubGet Early Access

The Repository

One monorepo. Everything from the CLI to the native ML runtime lives in a single repository with a unified build.

95+

Maven Modules

Apache 2.0

License

Java 17

Primary Language

4

Languages (Java, Rust, C++, Python)

Stack Architecture

Kompile is organized in layers. Each layer is a collection of modules you can build and run independently.

1

CLI & Orchestration

Single command-line harness that bootstraps, configures, and runs the entire stack. 100+ native commands. Passthrough mode for existing CLI agents.

kompile-clikompile-app-clikompile-model-clikompile-agent-clikompile-component-clikompile-cli-plugin-apikompile-cli-common
2

Application Layer

Modular application framework with swappable providers for LLMs, embeddings, vector stores, data sources, guardrails, and agents.

kompile-app (40+ sub-modules)kompile-app-corekompile-rag-pipelinekompile-guardrailskompile-graph-neo4jkompile-query-transformer
3

Model Server & Staging

Download, convert, optimize, train, and serve models locally. Multi-GPU scheduling, graph optimization pipeline, KV cache management.

kompile-model-stagingkompile-model-managerkompile-sdk-servingkompile-pipelines-framework
4

Model Importers

Convert models from any major framework into the native runtime format with automatic graph optimization.

kompile-model-importer-onnxkompile-model-importer-tensorflowkompile-model-importer-kerasGGUF/GGML ImporterSafeTensors Importer
5

ML Runtime

Native high-performance ML runtime with SameDiff automatic differentiation, ND4J tensor operations, and hardware-specific backends.

SameDiff (auto-diff engine)ND4J (tensor operations)LibND4J (native C++ backend)CUDA / cuDNN / Triton backendsCPU optimized (AVX, NEON)
6

Tokenizers & NLP

Rust-based HuggingFace tokenizer bindings and NLP utilities for text processing across the pipeline.

tokenizers-rustOpenNLP integrationSentence detection (30+ languages)
7

Projects & Knowledge

Self-contained project units that crawl data, build knowledge graphs, and compile structured knowledge.

kompile-project-storeKnowledge graph engineEntity extraction & resolutionCommunity detection

Technology Stack

The full set of technologies, frameworks, and libraries that make up Kompile.

Language

  • Java 17
  • Rust (tokenizers)
  • C++ (LibND4J)
  • Python (bridges)

Build

  • Apache Maven
  • Maven Wrapper
  • GraalVM Native Image

Frameworks

  • Spring Boot 3
  • Spring AI
  • Picocli (CLI)
  • Apache Camel

ML / DL

  • Eclipse DL4J
  • SameDiff
  • ND4J
  • Triton GPU compiler

Search

  • Anserini
  • Apache Lucene
  • BM25 + HNSW

Graph

  • Neo4j
  • Native graph engine
  • TransE / RotatE

Database

  • PostgreSQL
  • pgvector
  • Hibernate ORM

Model Formats

  • ONNX
  • TensorFlow
  • Keras
  • GGUF/GGML
  • SafeTensors

GPU

  • CUDA
  • cuDNN
  • Triton
  • Multi-GPU scheduling

Contribute

Kompile is built by a community of contributors. Whether you're adding a new data source connector, improving GPU scheduling, or fixing a bug in the pipeline — all contributions are welcome.

Star on GitHubOpen an Issue

Ready to compile your stack?

Get early access or dive into the source.

GitHubRequest Early Access