@Artedeingenio I like this style - it almost looks like digitized colored pencil, where the marks are still visible but some of the lines were turned into digital "vectors".
Very cool!
@intellectronica It's very good, you should give it a shot.
Biggest issue is the thinking time, and responses can take a while to come back.
But for what it is... being an open weight model... It's next level on that end.
@sam_paech@Zai_org From using it in the last day pretty significantly, it's a great model. Very capable...
Just wish they had a /fast mode like gpt-5.5 xhigh
At this point, I would just call it inefficient.
Code is not a means to an end, even if you enjoy it. Nobody types code to "relax", usually they're trying to build something.
And.. by extension these days... If you're trying to build something by writing code by hand...
= Inefficient.
@favelaoverlord Very useful for image and video still, yes.
Every word actually has to mean something though, and be there for a very specific reason.
"Fluff prompting" is useless.
@gmoneyNFT Yeah I agree ... Just feels "smooth" and something different from the other OS models, which I can only describe as sometimes "clunky" usage.
It's hard to put into words but I know exactly what you mean..
@openart_ai Very interesting! Is this more about automating the prompting, realism, scene-creation, and world-building?
Or does it factor in the storytelling side as well? And building up the actual concept from start to finish?
@JSFILMZ0412 It's definitely better than I expected... have they improved it more since the official announcement / API?
I was testing it through the API earlier and... this just seems better? Maybe it's just me.
Either way, solid video model...
Very well done! @heavypulp
And a great explanation overall.
More people need to hear about how tools like these "enhance" the creativity of the human behind the "virtual camera" (i.e. AI is essentially a renderer/camera now), rather than thinking it's a "click a button, out pops a video" type of deal.
Excited to try out the new model for putting together some shorts!
I'm very pleasantly surprised by this model! And the video generations I'm getting are very good.
The strange thing is, I was testing (what was said to be Grok Imagine 1.5) yesterday via. an external API... and the videos I got didn't come near what I'm getting right now through Imagine.
Would recommend everyone give it a shot - the quality is significantly improved compared to past versions!