I spent $30k and 3 months RL post-training an anime video model.
This is only step 30 out of a planned 1000 step run.
All samples are local text-to-video with no reference image/audio. Since it's based on LTX-2.3, each output takes under a minute on a single GPU.
I'm 19 and a solo researcher. Most of the budget went into ablations, reward design, and trying different configurations before reaching this setup.
The run is still extremely early, but the results already look much better than I expected.
It's compute-limited, not idea-limited.
I'm starting a company to continue scaling this and build frontier stylized video models.
If you're an investor, compute partner, video team, or someone who wants to help build this, DMs are open.
Kill the Angel - Episode 2
I continue the story with a more action-driven episode. I wanted to flex my editing chops, as well as my direction skills.
Total, took me about 18 hours to complete this, much less than Episode 1 because all of the voices and assets were pre-made. All visuals generated by Seedance 2.0. And all Dreamina monthly credits were used.
Unfortunately, Episode 3 will have to wait a few weeks, as I am gearing towards a epic fight video, which requires experimentation.
Please let me know what you think!
Please note: All of the music is by Yoko Kanno, from her various albums. Her music fits very well with the themes.
@sepharitestudio your dm is closed so i have to post here.
i'm from mochi, a mobile app dedicated to AI anime (we're new but growing fast, find us on the app store and see the reviews - mochi(.)tv).
we'd like to collab with you. shoot me a dm
Yorhel has passed away
He was known for running vndb, a website he created back in 2007
The moderation team is working to preserve the website but cannot provide details at the moment