Will the new version include free AI integration (DeepSeek, Grok, Gemini, etc.)?

[manuelm]

Hello,

I know the new version of WYSIWYG Web Builder will be released soon, and I’d like to know if it will include any functionality to work with free artificial intelligence services.

For example, will it be possible to connect with platforms such as DeepSeek, Grok, Gemini, or similar ones to generate text, images, or other content directly within the editor?

I think this would be a great improvement for those of us who use the program in creative and marketing projects, since it would make content production much easier.

Does anyone know if this feature is planned for the upcoming release, or if there are future plans in that direction?

Thanks in advance for any information!

[Pablo]

The current version already has support for Gemini (Google AI Studio).

Also, it supports free AI like deepseek, grok etc via Ollama.
https://www.wysiwygwebbuilder.com/ollama.html

And free image generation via ComfyUI integration (Flux, Qwen-Image, Z-Image etc).
https://www.wysiwygwebbuilder.com/comfyui.html

[manuelm]

Please note that the last two options (Ollama and ComfyUI) require local installation on a powerful computer in order to run properly.

[Pablo]

Correct.
But I am not aware of any other ways to get AI for free.
AI needs a powerful computer.

[wwonderfull]

@pablo is right. Because now a days people prefer offline solutions. On low-end pc it will take time but possible and on high-end pc it will do it faster, that is the difference.

Or the same thing you would have to rent from 3rd party provider who must process the same thing with a high-end pc in order to provide you the same service where you would need internet all the time, yet the asset belongs to their server which has less privacy and in many cases they own the assets generated.

So, the conclusion could be that self-sufficiency might be better for privacy and productivity.

[alan_sh]

I inslalled ComfyUI last year when V20 came out. I've got a powerful PC. It was OK, but quite limiting in what it produced and, for me, wasn't something I'd use every day.

Alan

[Pablo]

The results of ComfyUI depends on the selected model. Current models like Flux2. Z-image and qwen-image are producing very good results. Comparable with chat-gpt and other high-end models.
Also, z-image and flux2 klein are very fast.

[alan_sh]

I haven't updated it since V20 came out. Maybe I need to revisit it.

Alan

[Pablo]

There are new models released almost every week, so it's difficult to keep up.
But the latest are very good and also much faster (z-image and flux klein).