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

# Wally Cloud

> Hosted, OpenAI-compatible inference.

Cloud models run on RunAnywhere's hardware instead of the user's, so you can use models too
large for a laptop while keeping the same tools. They bill against prepaid credits, and they
are a separate path from on-device inference: nothing routes to the cloud on its own.

## Signing in

Cloud models need an account on [console.runanywhere.ai](https://console.runanywhere.ai). To
sign in, run:

```bash theme={null}
wally account login
```

The browser opens, you approve the request, and the terminal receives its credential. Check it
with `wally account whoami`, and see what you have spent with `wally account usage`.

## Running a cloud model

Hosted and local are separate choices, so ask for the cloud one:

```bash theme={null}
wally opencode --cloud --model <model-id>
```

`--cloud` is an `opencode` flag: it asks for the hosted model rather than one on this machine.
Set the model a harness uses by default with `wally models default`, and see
[Integrations](/integrations) for the tools Wally can point at a model.

## API access

The endpoint is OpenAI-compatible, so any OpenAI client works against it. Create a Cloud key
in the console, then:

```bash theme={null}
export RUNA_CLOUD_KEY='sk-runa-…'

curl https://inference.runanywhere.ai/v1/models \
  -H "Authorization: Bearer $RUNA_CLOUD_KEY"
```

That response is the authoritative list of models you can call. For request fields, streaming,
and errors, see the [API reference](/wally-cloud/api).