Download Center
Every desktop client in one place — pick your product and download for Windows.
AvatarHub Digital Human Engine
AvatarHub real-time digital human engine: voice cloning, live face swap, digital-human streaming and interpreting — deployed locally, data stays on-prem.
The Mac build is a lightweight console: heavy inference (face swap, digital human) runs on a Windows / server GPU; the Mac connects remotely.
Real multi-language output from one cloned voice — no voice actors.
Drag the slider to see the clarity the native face channel delivers.
Takes about 10–30 minutes depending on bandwidth, zero command line. Hover / tap dotted terms for explanations. See the manual for full feature guides.
- OS: Windows 10/11 (64-bit)
- GPU: NVIDIA 4 GB VRAM and up (8 GB+ recommended)
- Disk: 80 GB free on SSD
- Network: internet needed for first-time component download
- 1Download the installer~1 min
Click the download button on this page to get AvatarHub-Setup-x.x.x.exe (a ~60 MB ; AI components download on demand later).
- Optionally verify the shown on this page: run certutil -hashfile <installer path> SHA256 in PowerShell and compare.
- 2Run the installer~1 min
Double-click the installer and follow the wizard. It installs per-user with no admin rights and lets you choose the folder. An AvatarHub icon appears on the desktop and Start menu when done.
Most common snag: the "Windows protected your PC" prompt — a routine notice for recently published apps, not a virus alert. Click More info → Run anyway; verify the first if you like. - 3First launch · hardware detection~2 min
Open AvatarHub. The detects your GPU and recommends a (Lite / Standard / Flagship) — zero command line.
- Lite (4–6 GB [[VRAM]]): voice cloning, photo / video face swap.
- Standard (8 GB+ VRAM, RTX 3060 and up): live face swap, digital-human streaming.
- Flagship (24 GB VRAM, RTX 4090 / 5090): real-time HD plus the full interpreting suite.
- 4Automatic component download~10–30 min (bandwidth)
Confirm the tier and the wizard downloads the required AI components and models — every file verified, all downloads . Expect 10–60 GB depending on tier; keep 80 GB free on an SSD. Feel free to explore the UI meanwhile.
Slow or interrupted? Just reopen AvatarHub — continues from the breakpoint and finished parts are never re-downloaded. On weak networks, switch to a steadier connection. - 5Activate or start the trial~1 min
Go to Settings → License, enter your order number for (the signed license is fetched automatically) — or just start the 3-day full trial (clock starts at your first successful output; converts to a free-forever tier afterwards), no sign-up required.
- 6Verify the install~5–10 min
Click Start All in the launcher and wait for the service lights to turn green; run the health check — all green means you're done. Then a quick : upload a clear 10-second voice sample in the console → clone → play. If you hear it, everything works.
Paste the exact error to the AI assistant — install, GPU and download issues answered instantly; hand off to a human anytime.
Recommended: NVIDIA GPU with 8 GB+ VRAM (live swap / digital human); 4 GB for voice-only. Hands-off? Contact @WJKJ2026 for the 99 USD remote install service.
Everything after install: voice cloning, live face swap, phone interpreting, updates and troubleshooting. Read online or print to PDF in one click.
Release notes
One-click in-app updates: a single click downloads and installs new versions — components and character data preserved.
- Installer is now 1.5.5. The default folder stays D:\AvatarHub (or your user folder when D: is short on space) and no longer reuses a folder picked last time.
- After install it shows in Settings → Apps as “无界幻境 BOUNDLESS Studio”, so you can uninstall it from Windows.
- Uninstall also removes downloaded models, runtimes, and any moved-disk data, instead of leaving tens of gigabytes behind.
- 8GB GPUs now default to the FaceX clip-factory edition (~13.5GB): photo/video swaps plus presenter clips. The Lite edition is down to ~5.8GB.
- The first-run wizard asks what you want to do — clips, voice, or live avatar — and picks a tier from your VRAM. Live intent on a 24GB card goes straight to HD flagship.
- The download page now shows what your GPU can run: pick VRAM and intent, see a recommended edition.
- On 8GB machines only face-swap is warmed at boot; lip-sync and TTS load on demand and idle engines yield VRAM. Missing features install later from Settings.
- Products now sell separately: VoiceX (from 19) / FaceX (from 29) / LiveX (from 99, Pro = full bundle). Licenses are issued per product line — the console shows the workspaces you bought and honest upgrade cards for the rest.
- New FaceX clip-factory edition: photo/video face swaps and avatar presenter clips on an 8GB GPU (offline render queue), with dubbing on commercially licensed engines.
- New optional add-ons in Settings: the singing & vocal workshop (covers + separation + denoise, ~7.5GB) and ACE-Step original songs (~6.4GB) install with one click — keeping the base install light; source extraction upgrades automatically.
- Trial and free-tier policy unchanged: 3-day full trial on every product (clock starts at your first successful output), then the free-forever tier. Existing customers and licenses carry over seamlessly.