99.9% of what I see made w/ H3 on X is absolute slop which makes kino work like this by Maurice stand out even more
All generated with just 12GB of VRAM!
Minimax H3 running locally.
~65GB of models.
12GB VRAM. RTX 4070 Ti.
100% Text-to-Video.
100% ComfyUI.
Pass 1 → 960×544
Pass 2 → 1920×1088
One continuous loop built from multiple context-connected generations.
No cloud. No API.
CONTROL IS THE KEY.
LOCAL IS THE WAY.
We’re nearing the culmination of OAI’s redemption art
Next, I’m looking forward to their increasingly obnoxious and anti-competitive behaviour which will lead into Anthropic’s redemption arc
And so on and so forth until the end of time
See the best @MiniMax_AI H3 workflows in action as we announce winners of our #ComfyH3 Sync Sound challenge on a special livestream with guest judges @peterom, @emmacatnip, @yachimat_manga!
Tune in Thurs 9/3 at 10am PT
Most in the thread below believe that lacklustre Fable demand is due to Anthropic’s ZDR policy
My pet theory is that most economically useful work falls below a raw intelligence level + Fable is on the edge of this
The equivalent graph for Astra will be v. consequential
Lots of folks jumping to conclusions on the FT market share data for Fable. The reason is not at all obvious to me. Is it due to cost? Speed? Aggressive moves from OAI? ZDR? Open Source? Bad press / vibes? Refusal to work in some domains? Biased data?
