Skip to content

Draft to Take v3.0.0 beta 10

Choose a tag to compare

@JaySpiffy JaySpiffy released this 27 May 22:02
· 18 commits to main since this release

Draft to Take beta 10 is the current public beta.

Recommended Windows path:

  • Download DraftToTake-Native-Setup-v3.0.0-beta.10.exe from the assets below.
  • Run the installer and choose Full Studio (recommended).
  • Start Draft to Take from the Start Menu.

Installer notes:

  • This is a Windows installer preview, not a final signed production installer.
  • The installer is unsigned, so Windows SmartScreen may warn on first run.
  • NVIDIA GPU is strongly recommended.
  • Model files are not bundled in the setup exe; they download into %USERPROFILE%\DraftToTake\shared.
  • SHA256: 06AD65A37CEFC2A6F69C815D9EB32688FE8E3EC2B7FCCA4072C01A04F48A2F20

Docker fallback:

  • Docker users can still download the Source code zip or clone the repo.
  • The Docker launcher pulls public GHCR images tagged v3.0.0-beta.10.
  • Start with start.bat if you prefer the container path.

What changed:

  • Improves Script Canvas onboarding around prepared voices and the Voice Workbench.
  • Adds first-run starter voice guidance and clearer role-to-voice assignment.
  • Improves dark-mode readability, voice preview loading feedback, timeline/source editor resizing, and Script Canvas status messaging.
  • Refreshes public docs for the prepared-voice workflow and adds a Script Canvas AI system prompt.