10x National Security has open-sourced Nexus, a unified AI gateway that helps enterprises govern OpenAI, Anthropic and future LLMs through a single control plane while inviting the community to extend its capabilities.
10x National Security has open-sourced Nexus, an enterprise AI gateway and large language model (LLM) control plane designed to help organisations adopt AI while maintaining governance, security, observability, auditing and cost control. The project is now publicly available on GitLab.
Nexus provides a single governed control point for enterprise AI deployments, enabling organisations to manage AI model access, policies, routing, audit trails, observability and costs without tying governance to a single model provider. It currently supports OpenAI and Anthropic, with an architecture designed to accommodate additional providers and community-contributed integrations over time.
Developers can continue using OpenAI-compatible APIs or native Anthropic SDKs, requiring only a change to the base URL and API key rather than rewriting application code.
“AI adoption is moving faster than most organizations’ governance models,” said John Eubank IV, Founder and CEO of 10x National Security. “Teams are experimenting with models, agents, copilots, and AI applications across the enterprise, but too often they are doing it through direct provider calls, scattered API keys, disconnected logs, limited cost visibility, and fragmented policy enforcement. Nexus gives developers the speed they need while giving operators, security teams, and mission leaders the control point they need.”
The company is inviting developers, platform engineers, DevSecOps teams and security professionals to contribute provider integrations, OpenTelemetry support, SDK compatibility improvements, documentation, deployment examples, Helm and Terraform resources, and security enhancements.
Built using Rust microservices with a Next.js console, Nexus also offers provider routing, automatic retries, fallback between providers and models, optional response caching, policy enforcement, built-in cost tracking, observability, multi-tenant management, and support for bring-your-own-key workflows. Its open-source architecture is intended to provide organisations with a vendor-neutral foundation for deploying and governing AI at scale.
