Less than two years later, this is the quality of the video of an otter using a laptop on an airplane that I get (with sound) from an open weights AI video generator, MiniMax H3, running entirely on my local computer.
It took about 3 minutes to generate. Rapid progress, indeed.
Here’s Part 2, this time made with Seedance 2.5.
This is the final cut, built from shorter generations, and I’m really happy with how it came together.
My first attempt was one continuous 30-second generation, but it introduced too many mistakes. At 2.5’s price, repeatedly rerolling a long clip just wasn’t practical.
Breaking the sequence into shorter shots gave me much more control over blocking, character placement, vehicles, pacing and the final edit.
The biggest upgrade is the rendering. Skin has more texture and looks far less plastic than it often did in 2.0.
Multimodal referencing is genuinely useful too. I used character, vehicle and environment sheets, but never needed more than seven references at once.
Is it worth the extra cost? Yes, if time matters. I think 2.0 could reach a similar result, but it would take more generations and a lot more work.
A few example prompts in next posts 👇
Made via @dreamina_ai
I asked ChatGPT to show me a visual representation of the most insane phobia to have and apparently they think it's surveillance ducks...
Look at the reflection in the water 😭
Oh mi amor, mi amor, mi amor
Le cœur est froid, froid comme à la morgue
J’vois tout en noir comme, comme à la mort
Tout s’arrêtera le jour où toi et moi on se frotte