Snap allows you to generate 3D heads by text or image. Understand where automation helps and where alignment, entitlement, and compatibility require review.
Direct answer
Lens Studio 5.17's Head Generator creates a 3D head from text or image, presents four previews and imports the result into the project. Snap indicates that visual reference tends to offer better quality and that generation can take up to an hour. The feature speeds up avatar prototypes, but compatibility, facial alignment, similarity, consent and optimization need to be verified before public use.
Image improves direction, increases responsibility
Visual reference offers control, but requires clear authorization when representing a person.
Preview is not final asset
Eyes, mouth, profile and materials must be evaluated in expression and movement.
Alignment influences tracking
Position and scale adjustments prevent slippage and deformation during the experiment.
Compatibility must be confirmed
Documented limitations may prevent combination with certain features or pipelines.
Nexus Reading
Use generation in ideation, record source, and QA identity, animation, and device before publishing.
FAQ
Does Head Generator accept text and images?
Yes, Snap documents both input modes.
How many previews appear?
Four variations can be evaluated before importing.
Can a photo be used without authorization?
It is not recommended; rights, consent and purpose must be clear.
Essential guides to delve deeper into the decision
This editorial analysis was produced by Nexus from the official sources below, consulted on October 4, 2026. The text is original and interprets practical implications for companies.
- Snap — Head Morph Generation: inputs, previews, generation, import and limitations.
- Snap — Lens Studio detailed changelog: version context and official changes.
Date reported by main source: Lens Studio 5.17, documentation verified on October 4, 2026.