OpenAI people running these prompts have access to way more resources than any of us will through the API.

Cool, but see the drastic difference in quality ;)

Lack of quality in the details yes but the fact that characters and scenes depict consistent and real movement and evolution as opposed to the cinemagraph and frame morphing stuff we have had so far is still remarkable!

It has a comedy like quality lol

But all to be said, it is no less impressive after this new demo

There are absolutely example videos on their website which have worse quality than that.

Depends on the quality of the prompts.

Drastic difference in quality of the prompts too. Ones used in the OP are quite detailed ones mostly.

That particular example seems to have more a "cheap 3d" style to it but the actual synthesis seems on par with the examples. If the prompt had specified a different style it'd have that style instead. This kind of generation isn't like actual animating, "cheap 3d" style and "realistic cinematic" style take roughly the same amount of work to look right.

The output speed doesn't disprove possible cherry-picking, especially with batch generation.

Who cares? If it can be generated in 15 minutes then it's commercially useful.

Especially of you think that after you can get feedback and try again..15 minutes later have a new one...try again...etc

What is your point? That they make multiple ones and pick out the best ones? Well duh? That’s literally how the model is going to be used.

Please make your substantive points without swipes. This is in the site guidelines: https://news.ycombinator.com/newsguidelines.html.

Looks ready for _Wishbone_