Where is the journey headed?

Currently I am aiming for:

  • Running text generation models in a Docker container on the PC with the AMD GPU with Oobabooga textgeneration webui
  • Running image generation models in a Docker container on the PC with the Nvidia GPU with Automatic1111 Stable Diffusion WebUI
  • (later) Running image generation models in a Docker container on the PC with the Nvidia GPU with ComfyUI
  • Running Sillytavern in a Docker container on the PC with the Nvidia GPU (though the GPU should not matter in this case)
  • Running Silly-Tavern-Extras in a Docker container on the PC with the Nvidia GPU (it seemed not to work with AMD)
  • Running a text-to-speech AI on the PC with the Nvidia GPU with mit Alltalk-TTS. It seems that there is no Docker image for this yet. Maybe I can manage to “dockerize” this!
  • extract voices resp. voice models from various sources
  • (later) “Pimp” the text-to-speech AI to a realtime voice cloning (RVC) AI
  • extract voices resp. voice models from various sources
  • Make all this accessible in the LAN (reverse proxy on the NAS, VPN)
  • Create and maintain this website for keeping my insights, installation guides and tutorials so that I can have them at hand for the next installation... And so that others can also benefit from my insights.

And as has been shown in many places, the motto holds true: “The journey is the destination”! There are numerous pit stops along the way that, although not always necessary, make the “journey” more comfortable. For example, installing a proxy server, since I feel like I've installed Linux in all its variants and versions about a hundred times, and each installation of the AI programs resulted in the sometimes quite large libraries, like pytorch, being downloaded anew.

Further goals will become apparent on the way!