Regarding Image Generation

[Jeffser]

Hi everyone, I wanted to write an official statement regarding image generation implementations in Alpaca.

I've said before multiple times that I did not plan on adding it because of my own believes in regards to art. But recently I've come around to this.

Truth is, big companies have been gaining more and more control over the digital lives of people.

I believe in the importance of software alternatives, the same way Linux is an alternative to Windows and MacOS to a lot of people, there is a need for projects like OpenWebUI, Newelle and Alpaca as an alternative to online LLM providers.

Honestly I still don't like image generation, I come from a family full of artists and I learned from a young age to appreciate art as something special and that shouldn't be mass produced, believe me when I say I will never generate an image outside of testing for Alpaca.

But, I also understand that there's a big amount of people that make use of this technology and would appreciate it being integrated into an application they already use, in this case Alpaca, there's also the fact that Ollama is testing support for local image generation making it easier for me to integrate it.

I'm still unsure if I plan to add support for generating images using OpenAI instances, I really would prefer to concentrate my efforts in Ollama integrations, but that's something to be decided later.

---

So here's the deal, I will integrate image generation in a way that it only appears in the UI if the user has added an "Image Generation" model.

The idea is to have an Activity where you can select a model and prompt an image generation, there will also be a tool that creates a prompt based on the chat and calls the Activity to do the heavy lifting (similar to background remover).