> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zeroruntime.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Deployments
> How a Zero Runtime agent runs in production, and how to choose between Zero Runtime Cloud and self-hosting.
Your agent runs as a **worker**: your code, wrapped by `zeroruntime.serve(...)`. The Zero Runtime
**runtime** handles the real-time media (transport, GPUs, STT/LLM/TTS orchestration, and
turn-taking). Your worker registers with the runtime and receives sessions, so you never
operate low-latency media infrastructure.
## How a session flows
1. A **caller** starts a live session from a playground, a web or mobile client, or an
inbound phone call.
2. The session reaches the **Zero Runtime runtime**, which manages transport, GPUs, and
real-time media.
3. The **control plane** invokes your agent to start the session.
4. The runtime streams the session to one of your **registered workers**, which runs your
agent code and responds through the runtime.
## Choose where your worker runs
The runtime is always managed by Zero Runtime. You only choose who runs the worker.
| | Zero Runtime Cloud | Self-hosting |
| - | - | - |
| Who runs the worker | Zero Runtime | You (Docker, VM, Kubernetes) |
| Deploy with | `zeroruntime up` | Your own platform |
| Versions, secrets, logs | `zeroruntime` CLI | Your own tooling |
| Best for | Fastest path to production | Existing infrastructure, compliance, custom networking |
## Where to go next
Install the CLI and sign in.
Get an example agent running locally in about a minute.
Deploy your agent with one command.
Run the worker on your own infrastructure.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.