Solo.io, a provider of AI-ready infrastructure, has announced the general availability of its enterprise releases of Agent Substrate and kagent. The new offerings are designed to help enterprises build, deploy, and manage AI agents at scale with infrastructure-level isolation, security, and governance.
Agent Substrate introduces an infrastructure layer for agentic runtimes, combining serverless scaling, sandbox isolation, and governed connectivity. Meanwhile, kagent provides a developer experience for building and managing agents, tools, and skills. Launch customers, including OutSystems and Swissquote, are already running the infrastructure in production.
Enterprise GA releases: Solo.io has made enterprise versions of Agent Substrate and kagent generally available.
Agent Substrate: Introduces an infrastructure layer for agentic workloads, supporting serverless scaling, sandbox isolation, and governed connectivity.
kagent: Provides tools for building and managing AI agents, tools, and skills, with enterprise-grade monitoring and access controls.
Kubernetes integration: The platform introduces an execution model that suspends idle agent sessions and resumes them on demand.
Production deployments: OutSystems and Swissquote are using the infrastructure to support enterprise agent deployments.
Open-source momentum: Agent Substrate 1.0 has attracted more than 6,000 GitHub stars and over 100 contributors from more than 40 organizations.
As AI agents become a new category of enterprise workload, organizations need infrastructure that can support large numbers of sessions while maintaining security, performance, and operational control.
Kubernetes has become a widely adopted platform for managing traditional application workloads. However, agentic applications introduce different infrastructure requirements because individual agent sessions maintain their own state and context across multiple interactions.
At scale, assigning a dedicated pod to every agent session can consume significant computing resources, particularly when agents are waiting for model responses, tool execution, or human input.
Agent Substrate addresses this challenge through an execution model that suspends idle sessions and resumes them when needed. This allows organizations to scale agent workloads on their existing Kubernetes infrastructure without maintaining always-on compute resources for every session.
Agent Substrate introduces the actor as the fundamental runtime unit for AI agents. Each actor combines an agent session with two key infrastructure controls:
Sandbox isolation: Separates agent execution to help prevent one session from affecting another.
Governed connectivity: Uses a gateway to control who can access an agent and which resources it can reach.
The actor model keeps state, identity, and access controls associated with the agent session even as the underlying computing resources change.
This approach makes isolation and governance part of the runtime infrastructure rather than controls that organizations need to manage separately for each agent deployment.
“Agent Substrate is trying to do for agentic workloads what kubernetes did for traditional deployments - to make the orchestration of them a non-issue. This layer is necessary but it's not interesting for most agentic apps. Our goal is to make the problem go away, so people can focus on more interesting problems. Kubernetes itself is the perfect platform to build up from - it's everywhere and it's very well understood. When it comes to critical infrastructure, open always wins,” said Tim Hockin, Distinguished Software Engineer Kubernetes and GKE, Google.
Solo.io's enterprise distributions of Agent Substrate and kagent provide a supported runtime environment for agentic applications.
The platform uses agentgateway as the actor's gateway, enforcing identity and policy controls on incoming and outgoing connections.
Meanwhile, kagent provides the developer experience for building and managing agents, tools, and skills. Its enterprise capabilities include:
Granular access controls for managing permissions.
Agent tracing to help developers understand agent execution.
Infrastructure monitoring for operational visibility.
Evaluations built from real execution traces to assess agent behavior.
Solo.io also offers enterprise support and multiple deployment options, including air-gapped installations. These capabilities are designed for organizations operating in environments with strict security, compliance, and operational requirements.
Solo.io has worked with engineering teams at OutSystems and Swissquote to bring agentic infrastructure into production, addressing the scalability, security, and governance requirements of enterprise deployments.
OutSystems operates a large multi-tenant platform where customers require enterprise-grade resilience and scalability. The company identified always-on pods for individual agents as a challenge for supporting its expected workload.
Agent Substrate provides sandboxed runtimes that can suspend when idle and resume on demand. kagent and agentgateway add APIs, lifecycle management, and gateway controls to help integrate agents into the platform.
“We operate a large, multi-tenant platform where our customers expect enterprise-grade resilience and scale. With the number of agents we expect to support, running a dedicated, always-on pod for every agent simply isn't possible. Agent Substrate gives us sandboxed runtimes that suspend when idle and resume on demand. kagent and agentgateway build on that foundation with the APIs, lifecycle management, and gateway controls we need to integrate agents into our platform. Our partnership with Solo's team has been critical to bringing Agent Substrate, kagent, and agentgateway into production on our timeline,” said Ebenezer Schubert, VP Engineering, OutSystems.
Swissquote operates a sovereign AI stack that provides control over inference, infrastructure, and the AI lifecycle. The company is using Solo.io's enterprise kagent and Agent Substrate to provide infrastructure-level isolation and security while maintaining low latency for sensitive AI agents.
Its Codename Diego financial-crime investigation agents and MCP servers are already running in production on kagent GA.
“We operate a sovereign AI stack, with full control over inference, infrastructure and the AI lifecycle. Solo’s enterprise kagent and Agent Substrate provide infrastructure-level isolation and security, while delivering the low latency required by our most sensitive agents. Our Codename Diego financial-crime investigation agents and MCP servers are already running in production on kagent GA. In a multi-jurisdictional regulated environment, security, resilience and explainability are non-negotiable. The joint engineering work with Solo has been outstanding,” said Andrew Boswell, Head Information Technology, Swissquote.
Solo.io is a co-maintainer of the Agent Substrate project alongside Google and is contributing to the development of its open-source foundation for agentic workloads.
The community release of Agent Substrate 1.0 marks a milestone for the project, following six months of growth that brought more than 6,000 GitHub stars and over 100 contributors from more than 40 organizations.
To help developers and enterprise customers prepare for production deployments, Solo.io has also introduced a sizing calculator on the kagent project site. Based on the company's performance and scalability testing for general availability, the tool helps users estimate infrastructure requirements and establish service-level objectives for running Agent Substrate at scale.
“The next generation of applications will be agents. The infrastructure they need is being defined right now, and the choices we make will shape how enterprises run for years to come. Agent Substrate and the actor model are the foundation for that shift. We’re building for a world where agents carry out work across every part of an organization, with isolation and control built into the infrastructure from the start. Our customers are already showing what that future looks like in production,” said Idit Levine, founder and CEO of Solo.io.
Organizations and developers can explore Solo.io's latest agentic infrastructure offerings through the following resources:
Technical announcement: Announcing kagent 1.0 GA
Enterprise offerings: Solo.io Agent Substrate and kagent
Agent Substrate project: ate.dev
kagent project: kagent.dev
Agent Substrate sizing calculator: kagent.dev/substrate-calculator
With Agent Substrate and kagent now generally available in enterprise editions, Solo.io is positioning its infrastructure to support the growing demands of agentic applications. The combination of scalable execution, sandbox isolation, governed connectivity, and developer tools aims to help enterprises move AI agents from experimentation into production.
