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

# Concurrency

> Run Laravel Copilot SDK requests concurrently with Laravel Concurrency in stdio and TCP modes.

## Parallel execution with Laravel Concurrency

## Stdio mode

Starting a process per task is not efficient, but this pattern works reliably.

Because Concurrency serializes internal values, return a plain `content()` string instead of returning `SessionEvent`.

```php theme={null}
use Revolution\Copilot\Facades\Copilot;
use Illuminate\Support\Facades\Concurrency;

$prompt = 'Tell me something about Copilot.';

[$gpt5_response, $sonnet_response] = Concurrency::run([
    fn () => Copilot::run($prompt, config: ['model' => 'gpt-5.2'])->content(),
    fn () => Copilot::run($prompt, config: ['model' => 'claude-sonnet-4.5'])->content(),
]);

echo 'GPT-5 Response: '.$gpt5_response;
echo 'Claude Sonnet Response: '.$sonnet_response;
```

## TCP mode

Since there is no process boot per request, TCP mode can look faster, but observed latency may still be similar.

With the process driver, timeouts around 60 seconds are common in either mode.
With the CLI-only `fork` driver, one request may succeed while another fails.
In observed tests, switching from `gpt-5` to `gpt-5.2` allowed both requests to complete.

Concurrency exists as a feature, but queue-based processing is still the safer default for long-running workloads.

<Info>
  For the latest updates, see the [GitHub repository](https://github.com/invokable/laravel-copilot-sdk).
</Info>


## Related topics

- [Concurrency](/en/concurrency.md)
- [Upgrade guide: Laravel 11 to 12](/en/blog/upgrade-11-to-12.md)
- [HTTP client](/en/http-client.md)
- [April 2026 Laravel Updates](/en/blog/changelog/202604.md)
- [Guide to Building Apps with VOICEVOX Engine API](/en/packages/laravel-voicevox/app-guide.md)