HubLensLLMonyx-dot-app/onyx
// archived 2026-04-06
297

// summary

Onyx is a feature-rich open source AI platform designed to provide an easy-to-deploy application layer interface for large language models. The platform supports RAG, deep research, code execution, and various AI agent capabilities, while remaining compatible with mainstream self-hosted and proprietary LLMs. Users can deploy via the standard or lightweight versions to meet different needs ranging from personal use to enterprise-level collaboration.

// technical analysis

Onyx is a full-featured open-source AI platform designed to serve as an application layer for Large Language Models (LLMs), providing users with a self-hostable interactive interface. By integrating RAG, deep research, code execution, and multimodal capabilities, the project solves integration challenges for enterprises and individuals when using LLMs for complex tasks. Its architecture is flexible, supporting both standard and Lite deployment modes to accommodate varying resource requirements from personal testing to large-scale enterprise applications, and it provides extensive connectors and MCP support to expand external application interaction.

// key highlights

01
Agentic RAG combines hybrid indexing with AI agents, significantly improving the accuracy of information retrieval and the quality of responses.
02
The deep research feature supports multi-step research workflows, capable of generating detailed analytical reports.
03
A built-in code execution sandbox allows users to perform data analysis, graphical rendering, and file modifications directly within the platform.
04
Supports over 50 out-of-the-box indexing connectors and the MCP protocol, making it easy to connect to external applications.
05
Provides voice mode, image generation, and custom AI agent building to meet diverse interaction and creation needs.
06
Enterprise-grade features include SSO, RBAC permission control, and detailed usage analytics, ensuring secure and compliant AI deployment within organizations.

// use cases

01
Utilize Agentic RAG and deep research features for efficient information retrieval and analysis
02
Interact with external applications through custom AI agents and the MCP protocol
03
Execute code in a sandbox environment for data analysis, graphical rendering, or file processing

// getting started

Developers can perform a one-click deployment by executing the command 'curl -fsSL https://onyx.app/install_onyx.sh | bash'. For resource-constrained environments, you can choose to deploy the Onyx Lite mode, or refer to the official documentation for detailed deployment guides for Docker, Kubernetes, and cloud platforms.