# Google AX

> Updated 2026-09-21 · type: tool · category: ai-infrastructure · status: active · rev 1

Google AX is an Apache-2.0 orchestrator that runs declarative agent tasks in sandboxed, network-fenced workspaces at scale on Kubernetes.

- Open source: yes (Apache-2.0)
- Self-hostable: yes
- Pricing model: free
- Best for: A team that has moved past one agent on a laptop and needs to run many agent tasks reliably, each sandboxed and network-fenced, on infrastructure it controls, rather than trusting an agent loose on a shared machine.
- Last verified: 2026-09-21

- **Canonical:** https://gtmstacker.com/registry/tool/google-ax/
- **Source:** [github · google/ax](https://github.com/google/ax)
- **Tags:** ai-infrastructure, mcp-agents, orchestration, self-hostable, sandboxing
- **Repository:** https://github.com/google/ax

## Is Google AX open source?

Yes, Google AX is open source under the Apache-2.0 license.

## How much does Google AX cost?

Google AX is free to use.

## Can I self-host Google AX?

Yes, Google AX can be self-hosted (the source is available under the Apache-2.0 license).

## Alternatives & related

- [Octop](https://gtmstacker.com/registry/tool/octop/)
- [HarnessRouter](https://gtmstacker.com/registry/tool/harnessrouter/)


---

Apache-2.0 agentic orchestrator from Google (Go) for running autonomous agent workloads at scale. You declare an agent task in a YAML manifest, its workspace and its gateway, and AX sandboxes the task, wires up the workspace, fences its network, and runs it on Kubernetes through a control plane. It integrates with LLM providers you supply, so it handles isolation and scale-out while you bring the model. Early observers describe it as Kubernetes for agent executions.

## Provenance

- Apache-2.0, Go, ~4.8k stars, ~625 commits, and the declarative sandbox-and-fence orchestration model independently WebFetch-verified on the repo 2026-09-21 (github.com/google/ax).
- Surfaced via the 2026-09-21 viral-posts brief (a credible engineer's "Google open-sourced AX, feels like Kubernetes for agent executions").
- Curated from the GTM Stacker signal registry (2026-09-21 pass); license independently verified 2026-09-21.

## Why it matters for a GTM stack

Once agents graduate from a demo to doing real work across a team, the problem stops being the prompt and becomes operations: how do you run twenty agent tasks without one of them wandering out onto the network or trampling a shared machine. AX is Google's answer, declare the task, and it runs sandboxed and network-fenced, at scale, on infrastructure you own. For a GTM or RevOps team that wants agents touching real systems with a blast radius you can set, that isolation is the feature. The honest read: it assumes Kubernetes and a control plane, so this is platform work, not a weekend install, and you still bring and pay for the models. As the piece that makes running many agents an operational discipline instead of a hope, it is worth the pilot.