> ## Documentation Index
> Fetch the complete documentation index at: https://langchain-5e9cc07a-preview-docsfl-1776371397-e812a24.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Sandbox templates

> Define container images, resource limits, and configuration for sandboxes using templates.

<Warning>
  Sandboxes are in private preview. APIs and features may change as we iterate. [Sign up for the waitlist](https://www.langchain.com/langsmith-sandboxes-waitlist?ref=docs.langchain.com) to get access.
</Warning>

Before creating a sandbox, you need a **template**. Templates define the blueprint for sandbox instances, including the container image, resource allocation, and optional configuration like volumes and auth proxy rules.

<img src="https://mintcdn.com/langchain-5e9cc07a-preview-docsfl-1776371397-e812a24/g8BB-HNcnflnTrm7/images/langsmith/sandboxes/sb-templates.png?fit=max&auto=format&n=g8BB-HNcnflnTrm7&q=85&s=3fc11198ea4765fc0e79a531a2f44cf8" alt="Template list view" width="3206" height="942" data-path="images/langsmith/sandboxes/sb-templates.png" />

## What templates define

| Setting               | Description                                                                                                     |
| --------------------- | --------------------------------------------------------------------------------------------------------------- |
| **Container image**   | The Docker image to use (private registries are supported)                                                      |
| **Resource capacity** | CPU, memory, and storage limits                                                                                 |
| **Volumes**           | Persistent storage to attach (optional)                                                                         |
| **Auth proxy config** | Rules for injecting secrets into outbound requests (optional) — see [Auth proxy](/langsmith/sandbox-auth-proxy) |

## Create a template

<CodeGroup>
  ```python Python theme={null}
  from langsmith.sandbox import SandboxClient

  client = SandboxClient()

  client.create_template(
      name="python-sandbox",
      image="python:3.12-slim",
  )
  ```

  ```ts TypeScript theme={null}
  import { SandboxClient } from "langsmith/experimental/sandbox";

  const client = new SandboxClient();

  await client.createTemplate("node-sandbox", {
    image: "node:20-slim",
  });
  ```
</CodeGroup>

You can also create templates via the REST API with full control over resources and proxy configuration:

```bash theme={null}
curl -X POST "$LANGSMITH_ENDPOINT/api/v2/sandboxes/templates" \
  -H "x-api-key: $LANGSMITH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "python-sandbox",
    "image": "python:3.12-slim",
    "resources": {"cpu": "500m", "memory": "512Mi", "storage": "2Gi"}
  }'
```

## Next steps

* [Create a sandbox](/langsmith/sandbox-sdk) from your template
* [Configure auth proxy rules](/langsmith/sandbox-auth-proxy) to inject secrets into outbound requests
* [Set up warm pools](/langsmith/sandbox-warm-pools) to pre-provision sandboxes for faster startup

***

<div className="source-links">
  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/langsmith/sandbox-templates.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>

  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>
</div>