You need an Apple Silicon Mac, an available device slot in the intended Control account or organisation, and internet access. Allow disk space and memory for the model and its engine. The Aventail 2.0.0 package installs for your macOS user without an administrator prompt.
Tested installation and local API: Aventail Link 2.0.0 · macOS 26.6.2 · Llama 3.2 3B Instruct Q4_K_M. The separate production Control and Workspace captures use Locai Link (Legacy Aventail) 1.3.2.
If Aventail Link is already installed, confirm the device is Online in the intended Control environment and skip to deployment. Locai Link (Legacy Aventail) 1.x has a separate installation; see upgrading from the old app.
Open your organisation's Control environment and sign in. Production Control and development Control have separate accounts and devices. Use the same environment and account when Link asks you to sign in.
AVENTAIL / GET STARTED
Sign in to Control
1 / 3
Full view
Control sign-in page
The account controls are on the right of the sign-in page.
Find the section
Sign-in form
The outlined area is shown close up on the next slide.
Close-up
Choose your sign-in method
Open Control and use Google, GitHub or your email and password. Use the provider you originally registered with.
Slide 1 of 3: Control sign-in page.
Control sign-in interface. Use the arrows or progress marks to move between slides.First sign-in and account options
Choose Google, GitHub, or your email and password. On first sign-in, review and accept the Terms & Conditions if you agree, then close What's new. New accounts open the web setup wizard over the Dashboard.
Follow the walkthrough, then check your Mac is Online in Control.
Download Link for macOS (Apple Silicon). This is the production release. For a development or staging account, use the download offered by that Control environment.
AVENTAIL / LINK
Install and connect Aventail Link
1 / 6
Full view
Aventail Link installer
The installer shows its stages on the left and the current action on the right.
Find the section
Introduction and Continue
The outlined area is shown close up on the next slide.
Close-up
Open the package
Open the macOS ARM64 .pkg supplied for your Control environment, then choose Continue.
Slide 04
Install for your user
Choose Install. This package installs into your home folder and is available to your macOS user.
Slide 05
Finish the installer
Wait for The installation was successful, then choose Close. The separate Aventail Link Setup window opens.
Slide 06
Sign in to Aventail
Select Sign in with Aventail. In the browser, check the account and organisation before approving this device.
The captured 2.0.0 build opens development Control. Use the environment intended for your package.
Slide 1 of 6: Aventail Link installer.
Aventail Link 2.0.0 · Apple Silicon · captures from 2 October 2026. Use the arrows or progress marks to move between slides.Read the installation steps
Obtain the Aventail Link 2.0.0 macOS ARM64 .pkg intended for your Control environment. Use the production download above, or the download offered by your development Control environment.
Open the package, choose Continue → Install, wait for The installation was successful, then choose Close.
In Aventail Link Setup → Connect, select Sign in with Aventail. In your default browser, check the account and organisation, compare the approval code with Link and approve the device.
Return to the Models screen and continue below.
The 2.0.0 build used for these screenshots opens development Control during sign-in. The public download above is the production channel; keep the package and account environment together. It has no installer licence step or administrator prompt, and the package starts the user service automatically. Setup continues through Models → Serving → Finish.
If you downloaded and enabled a model during Aventail setup, it is already serving: continue to your first response. Otherwise, select a model in setup or deploy one from Control. Keep Link running while it downloads.
AVENTAIL / LINK
Download and serve during setup
1 / 6
Full view
Aventail model setup
After connecting, choose the models this device should download.
Find the section
Model catalogue and download sizes
The outlined area is shown close up on the next slide.
Close-up
Choose your models
Tick a model to start downloading it. You can leave all models unticked to deploy from Control later, then select Continue.
Available models depend on your account. This capture shows the development catalogue.
Slide 04
Let the download finish
Selecting a model starts its download immediately. Wait for it to finish before expecting a serving toggle on the next screen.
This example uses a 2.0 GB model from the development account.
Slide 05
Start serving
Once the download is complete, tick the installed model to serve it on port 8100, then choose Continue.
The image shows the model before its serving checkbox is turned on.
Slide 06
Finish setup
Check All set and the installed/serving counts, then select Finish at the bottom-right of the setup window.
Redacted copy of the supplied capture. This walkthrough finished with one model serving.
Slide 1 of 6: Aventail model setup.
Aventail Link 2.0.0 · development account · 2 October 2026. Identity fields redacted in the Finish image. Use the arrows or progress marks to move between slides.Read the deployment steps and Control alternative
In Models, tick a model to start downloading it. The captured 2.0.0 walkthrough uses llama_3.2-3B-Instruct (2.0 GB); your account catalogue can differ.
Wait for the download, then select Continue. In Serving, tick the installed model to serve it on port 8100. If Nothing downloaded yet appears, use Back to check progress.
Select Continue, check All set and the installed/serving counts, then select Finish. The captured result shows 1 installed · 1 serving.
The earlier production Control route is shown below. Use Control in the same environment as the registered device; the development fleet interface may differ.
AVENTAIL / GET STARTED
Deploy and start serving
1 / 5
Full view
Deployment in the setup wizard
After connecting the device, the wizard recommends a starter model.
Find the section
Deploy stage
The outlined area is shown close up on the next slide.
Close-up
Deploy the starter model
In Control’s wizard, choose Continue to deploy → Deploy model. Keep Link running while the model downloads.
The device name and available model can differ.
Slide 04
Start serving
Wait for deployment to complete. Select Serve Model, or turn on the model’s switch under Serving. Use an available port; this walkthrough uses 8100.
The completed deployment dialog is shown here. Wizard deployments also appear in Serving.
Slide 05
Check the endpoint
In Serving → API Usage, check the endpoint for your model. On this Mac, use 127.0.0.1 and the serving port.
Slide 1 of 5: Deployment in the setup wizard.
Control deployment and serving screens captured 28 September 2026. Use the arrows or progress marks to move between slides.
In the wizard, select Continue to deploy → Deploy model. Wait for Deployed in Notifications → Deployments, then open Serving, expand your device and turn on the model's switch. Enter port 8100, select Start Serving and wait for serving.
To deploy from Models instead:
Open Models → Globally Shared in Control. Find Gemma 4 (X-Small), the GGUF model used for the earlier production Control walkthrough (2.44 GB download).
Select Deploy, choose your Mac under Target Device, and select Deploy Model.
Keep Link running until Model deployed successfully! appears.
Keep the displayed port (the tested default was 8100) and select Serve Model. Confirm the model appears under Serving.
Deployment downloads and prepares the model. Serving starts it so applications can send requests. The first request may take longer while the model loads.
If the recommended model is unavailable or deployment fails, see deployment help.