InseeFrLab/onyxia

🔬 Data science environment for k8s

What it solves

Onyxia provides a centralized web interface for data scientists to deploy and manage their working environments on a Kubernetes cluster. It removes the need for data scientists to manually configure complex infrastructure, allowing them to launch a variety of data science tools (like Jupyter and RStudio) on demand.

How it works

Onyxia acts as a "glue" layer between various open-source backend technologies. It allows users to launch containers from a catalog of Helm charts, with the following capabilities:

  • Resource Allocation: Users can specify the amount of CPU, RAM, and GPU resources required for their session.
  • Customization: Users can define environment variables and custom initialization scripts to be executed at launch.
  • Integration: It integrates deeply with S3 for data storage, Vault for secret management, and Keycloak for authentication.

Who it’s for

Data scientists and organizations that use Kubernetes to provide scalable, on-demand working environments for their research and analysis.

Highlights

  • On-demand environments: Launch Jupyter, RStudio, and other tools via Helm charts.
  • GPU support: Direct control over GPU allocation for containers.
  • Enterprise integrations: Built-in support for S3, Vault, and Keycloak.
  • Configuration management: Ability to save and restore service configurations.

Related

  • Project
    odysseus-dev/odysseus

    A self-hosted AI workspace that integrates chat, agents, research, and document management with email, calendar, and notes for a unified productivity environment.

  • Project
    dell/omnia

    Omnia is Dell’s open‑source Ansible‑based toolkit for automatically provisioning Linux servers, installing drivers, and deploying Slurm or Kubernetes together with AI/ML frameworks. It streamlines the creation of HPC or on‑prem AI clusters, is Apache 2.0 licensed, and is actively maintained by a community that includes Dell, Intel, universities and other partners.

  • Project
    omnigent-ai/omnigent

    An open-source meta-harness that provides a common orchestration layer for multiple AI agents, enabling cross-device synchronization and cloud sandboxing.

  • Project
    lablup/backend.ai

    Backend.AI is an open‑source, container‑based cluster platform for running AI/ML workloads. It schedules isolated compute sessions (Jupyter, VS Code, SSH, etc.) on heterogeneous accelerators, offers a virtual‑folder storage layer, and exposes REST/GraphQL APIs plus Python/Java/JS SDKs. Install via a single script, start Manager/Agent/Storage/Web services, and launch sessions with a CLI or web UI. LGPL‑v3 server code, MIT client SDKs.

  • Project
    onyx-dot-app/onyx

    Onyx is an open‑source, self‑hosted platform that indexes data from 50+ workplace apps, builds a hybrid vector‑keyword store, and exposes that knowledge to LLMs via agents. It supports web search, sandboxed code execution, external actions, voice interaction, and works with any major LLM provider. Deployable via Docker/Kubernetes (standard or lightweight Lite mode) and offered under an MIT‑licensed Community Edition, with an Enterprise add‑on for larger teams.