Year: 2026
Type: article
Abstract: OGX (Open GenAI Stack), formerly Llama Stack (Llama Stack Contributors, 2025), is an open-source AI application server and Python library with interchangeable backend providers (OGX Contributors, 2026b).Teams building retrieval-augmented generation (RAG) pipelines, conversational agents, and tool-calling workflows can change their inference engine and vector database through server configuration without rewriting application code.OGX implements OpenAI, Anthropic, and Google APIs, with the Responses API for server-side orchestration as its primary focus (Open Responses Community, 2026).The cited v1.5.0 release supports configurable inference and vector storage providers.A companion Kubernetes Operator (OGX Contributors, 2026a) manages deployments, including shared and per-tenant instances.Together, OGX and its operator provide a self-hosted backend for AI-powered developer tools such as Claude Code, Codex CLI, OpenCode, and OpenHands.
DOI: 10.21105/joss.11234
Institutions: Red Hat (United States), Meta (United States)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7203679066