Disclaimer: I am a little frustrated with the censorship on Grok, the inability of other platforms to understand how a spit roast should look like and thus... some questions.
I won't ask about timelines, I guess it is done when it's done and I can be patient. Sometimes.

But what would be really interesting, if it can be shared at this point, is the scope of the project. My current udnerstanding is that this model will be specifically trained with out niche fetish in mind, so finally a "pot on a fire in a cannibal village" will be able to hold a human (or two) and not be either tiny or in reality a pan.
But... what will be the capabilities?
Text-To-Image?
Image-To-Image?
Text-To-Video?
Image-To-Video?
All of the above?
And if videos are in scope, does it include sound?
And finally: Will it just be "available" to all paying members or will there be limitations in place, like a Credit system (like many AI sites use) or a daily rate (like e.g. Grok uses it)?
Thanks for indulging me, keep up the good work, you are awesome!
