Hands-on: Meta’s Hologram Avatars—what they are, how they’re made, and where they’ll appear first

# Holograms

# How Holograms are created The setup differs by device. For Ray-Ban Display, you use Meta's Android or iOS app and your phone's selfie camera. The capture flow asks you to turn your head, smile, laugh, and speak short prompts (for example, answering "What's your favorite kind of day?"). Because Display's implementation is driven by voice, those vocal inputs matter: the AI uses them to animate and time expressions and speech.

Meta allows up to 10 tries per day for a capture, which staff said can be necessary to get a more convincing likeness. When you call someone and they're represented by a Hologram, a small ghostly Hologram icon appears in the call interface so you know you're talking to an avatar.

# Display versus VR Glasses: different hardware, different experiences Ray-Ban Display is a see-through HUD with a small single display in the right eye. Holograms there are shoulders-up and operate more like an enhanced flatscreen video call. Because the visible display area is limited, the experience reads as more useful for smartphone-style interactions than for augmenting in-person presence.

Meta VR Glasses add eye and face tracking and have downward-facing sensors that capture head-to-toe information. That allows Meta to produce a more realistic full-bodied avatar. The demo suggested the capture for the VR Glasses will include head-to-toe scanning and likely specific posed moves so the system can learn your neutral and more demonstrative gestures. The result in the demo was a convincingly realistic full-body avatar for some people.

# What the hands-on revealed about realism and limits

Because Display Holograms are primarily voice-driven and constrained to a small HUD, the effect is subtler. VR Glasses' sensors deliver more data for motion and eye behavior, so avatars can appear more lifelike there, especially when full-body capture succeeds.

# Availability and platforms

# Practical implications For users: expect a staged rollout that starts with shoulders-up, voice-driven avatars usable in phone-style calls, then expands to richer full-body telepresence on VR hardware with more sensors. Expect to retry captures if you want a closer likeness. App interfaces will make it clear when you're speaking to an avatar rather than a live person.

# Bottom line