> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tensormesh.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Playground

> Use the repository playground as a reference Next.js integration.

The public repository includes a BYOK Next.js playground under
`examples/playground`. It is reference code for application integration and is
not included in the npm package tarball.

It demonstrates:

* Streaming chat with `streamText`
* Structured output with Zod-backed schemas
* Tool calling with AI SDK tools
* Runtime model selection from the public `GET /v1/models` endpoint
* Serverless inference only
* Code samples for the active workflow

Clone the repository and install dependencies:

```bash theme={null}
git clone https://github.com/Tensormesh-Production/tensormesh-ai-sdk-provider.git
cd tensormesh-ai-sdk-provider/examples/playground
npm install
```

Run the playground locally:

```bash theme={null}
npm run dev
```

Open `http://localhost:3000` and paste a Tensormesh inference API key. The
playground does not store API keys; the browser sends the key to the app's API
routes for each request.

For Vercel deployment, import the repository and set the project root directory
to `examples/playground`. No environment variables are required because the
playground is BYOK.

See the [playground README](https://github.com/Tensormesh-Production/tensormesh-ai-sdk-provider/tree/main/examples/playground) for the full setup.