Ready Player Me was the default way to give a web app, game or VR world a customisable 3D avatar. In December 2025 Netflix announced it had acquired the company, and the public service closed on 31 January 2026: the avatar creator, the APIs and the developer tools (Wikipedia).
If your product used Ready Player Me, you are left with whatever you exported. The good news is that a glTF file doesn't need the platform to work.
What still works
- GLB files you downloaded and host yourself keep working: they are standard glTF.
- Links to Ready Player Me's servers do not, because the models were served from its CDN and created through its API. Copy every avatar you still need onto your own storage, and update the URLs in your app.
- New avatars can't be created with the Ready Player Me creator. You need another source for new characters, or your own pipeline.
Check what your exported avatars contain
Ready Player Me did not include facial blend shapes by default: you asked for them with the morphTargets parameter when downloading, for example morphTargets=ARKit,Oculus Visemes. So two avatars from the same account can contain different shapes. Check each one:
gltf.scene.traverse((o) => {
if (o.morphTargetDictionary) console.log(o.name, Object.keys(o.morphTargetDictionary));
});
What you might see on the head mesh (usually Wolf3D_Head), and on the teeth (Wolf3D_Teeth):
| You see | The avatar has | Lip sync options |
|---|---|---|
jawOpen, mouthFunnel, mouthPressLeft, ... |
ARKit blend shapes | ARKit curves (best) |
viseme_sil, viseme_PP, viseme_aa, ... |
Oculus visemes | Viseme timelines |
| Both | Both sets | Either |
Only mouthOpen, mouthSmile |
The default minimum | Volume-driven jaw only, or re-export elsewhere |
Make sure the teeth mesh carries the same shapes as the head; otherwise the jaw opens around a fixed set of teeth.
Giving them lip sync now
Ready Player Me never generated speech animation itself; projects combined its avatars with a separate lip sync source. That part is unaffected by the shutdown, and the choice depends on the shapes you found:
- ARKit avatars: drive them with ARKit curves, from text and audio. The playback code is in our three.js lip sync guide, and the shapes are explained in ARKit blend shapes explained.
- Oculus viseme avatars: you need viseme timings. Meta has also retired the Oculus Lipsync plugin; the alternatives are in Oculus Lipsync is end-of-life.
- Both: prefer ARKit; it describes the mouth in more detail and reads more naturally.
Planning for the next avatars
Two lessons from the shutdown:
- Own the files. Keep source models and exported GLBs in your own storage, never only behind a vendor's URL.
- Use a standard face. Characters with the 52 ARKit blend shapes work with most animation and lip sync tools, so you can change any one tool without redoing your characters. Character Creator and most custom pipelines can produce them, and VRM avatars can carry them as "Perfect Sync" shapes.
Our Lip sync API works with exported Ready Player Me avatars that have ARKit shapes: send the text and its audio, get curves for each frame, play them on the head and teeth meshes.