Aillex is a local-first AI companion. In her newest setup, a live talking face on the touchscreen of the PC case and on a phone, her ears and face run on an RTX 5090 gaming PC and her brain runs on an older server PC with two 2016 graphics cards, both in our house. Her apps normally speak with a local voice. On the live face, her voice comes from ElevenLabs temporarily, while it is fast and effectively free, because the local voice was too slow sharing the graphics card with her face. Every part is swappable, and the guides show a fully local version of each.

She’s also a demonstration: everything she can do, you can build. This site is the build manual, the real recipes, including the failures and the dependency hell, and the YouTube channel is where she shows it off (and hosts daily AI tips, presented by her sister persona Xellia).

What she can do today

  • Converse by voice from the touchscreen on the side of the PC case or from a phone, in the voice she was designed with (a local clone in her apps; ElevenLabs, temporarily, on the live face)
  • Talk back with a live face: one photo of her, animated in real time by SoulX-FlashHead Pro on the RTX 5090
  • Remember what matters: she keeps the whole conversation in context, and the important points get written into her personality and reference notes, which carry over
  • See, describe your screen on demand using the same local model
  • Join Discord voice calls, the same brain, another surface

The philosophy

Cloud companions can change their pricing, personality, or privacy policy overnight. A local companion is yours, which is exactly why we build in the open and teach every step. That’s why everything we can run at home runs at home. The one exception right now is her voice on the new live face, which comes from the cloud temporarily; her other apps, Discord included, still speak locally.

Follow the build