Device Access Toolkit v0.7 out today with Display! 🎉

[OscarFalmer]

Hi everyone,

We’re excited to announce the release of version 0.7 of the Meta Wearables Device Access Toolkit.
The wait is over, build for display glasses starting today!
Availability is currently rolling out starting today over the next couple weeks.

📌 Release Notes

• New Display capability: display visual experiences on Meta Ray-Ban Display’s glasses including images, texts, videos, buttons.
• New Device Access Toolkit App Model (DAM): a new architecture model for the SDK. DAM is required for the new Display capability. Both the App Model flow and the older flow continue to be supported for camera functionality on Meta AI glasses.
• New DeviceState & ThermalLevel values: device live state observation

📌 Related News

• WebApps on Meta Ray-Ban Display: starting today you can create Web Apps for the Display, learn more here in our announcement post and dedicated GitHub repo
• Display Recording: with the firmware rolling out starting today over the next couple weeks, you can also now record content on Display and share what you built with others.

🔗 Check the iOS changelog and Android changelog to see full changelogs.
🔗 Check the full Documentation on Wearables Developer Center.
🔗 Post your projects on the new Meta Subreddit for AI glasses developers.

[henryliang3027]

Hi Meta Wearables team,

Thanks for releasing the new SDK support for Display capabilities — it's exciting to see the ecosystem growing.

I would like to ask whether there are any plans to expose Neural Band gesture signals through the SDK in future releases.

For example:

onClick / tap gesture events
Camera zoom in / zoom out gestures
Gesture callbacks or event streams
Raw or processed Neural Band input APIs (if applicable)

One of my use cases is building an application that uses the glasses camera to capture text on snack packaging. Users often need to zoom in for better OCR accuracy and reduce unnecessary background content. Since users may be holding products or placing them on a desk while scanning, Neural Band gestures could provide a more natural interaction method than relying only on voice commands or phone interactions.

Is Neural Band input currently restricted to system-level interactions only, or is SDK support planned on the roadmap?

Currently, is Neural Band gesture input intentionally restricted to system-level interactions only, or is SDK support planned on the roadmap?

Would appreciate any guidance on future plans or recommended alternatives for developers interested in gesture-driven experiences.

Thanks!

[mikarinneoracle]

Hi! I converted my old app to this latest version, works well! Am I imagining it or is the video capture quality improved (I'm taking frames from the stream) - it looks much better now ? Or maybe it's just me ..