If you have any good tips and workflows for creating H3 videos, please feel free to message me. I will reward you! Please include your method, prompt, and your Group ID (not UID)!
@MiniMax_AI@Hailuo_AI
More context on H3’s U.S. licensing below.
And continuing the excitement for the open weight launch: H3 is verified to run on GPUs including the RTX 5090 and RTX 6000.
Check it here https://t.co/v0LEsmSWGd
This is exactly why we’re going open-weight — only open-weight lets us rally a broader community to build together. 🤗
@ComfyUI community’s quantized builds already support a wide range of GPUs, including the RTX 6000, dual 3090s, 5060, and 4060 Ti.
More GPU models are currently undergoing testing. Feel free to share any benchmark tests under this thread.
hello, world! i'm daniel! currently i'm the head of r&d of MiniMax code @MiniMax_AI, which is a frontier ai lab in china.
i enjoy working on our ai-native product @MiniMaxAgent . i'm really obsessed with this question: what does an ai-native organization look like? how can our organization co-evolve together with llms?
besides that, i'm passionate about open source. i'm the co-author of the deerflow project, which is a representative work in harness engineering and has 78k 🌟 on github.
if you're passionate about open source, foundation models, or agents- or if you enjoy playing basketball or riding a road bike - let's chat and be friends! :)
I’ve always believed superintelligence safety will not be solved by simply making AI “nice”.
The nuclear age taught us something important: stability often comes from game theory.
Nuclear weapons did not prevent catastrophe because humans became morally perfect. They prevented it because the incentives changed — launching first became the irrational move.
A future with super AI may require a similar mechanism:
Not an AI that is forced to obey.
But a system where cooperation is the Nash equilibrium, and betrayal is the losing strategy.
The biggest challenge of the AI era may not be intelligence itself.
It may be designing the game that intelligent agents will play.
hello there the jacobian conjecture is false thanx to my close friend akhil for asking about it and my other close friend fable for working during the world cup final
((1+xy)^3 z + y^2 (1+xy) (4+3xy), y + 3 x (1+xy)^2 z + 3 x y^2 (4+3xy), 2 x - 3 x^2 y - x^3 z): \C^3\to \C^3, has jacobian determinant -2, and sends (0, 0, -1/4), (1, -3/2, 13/2), and (-1, 3/2, 13/2) to (-1/4, 0, 0)
Too many teams treat every milestone as a miniature finish line, then optimize for short-term metrics.
A milestone should validate scalable methods, expose risks, and build reusable systems.
It doesn’t have to win; it has to make the final version more likely to win.
My first SIGGRAPH!
If you’re attending, let’s coffee and talk all things graphics, AI, and creativity
btw this was my graduation headshot almost 10yrs ago and the gray dress is now my precious 5-year-old pajama😆
Always happy to be back in LA��️ #ForeverTrojan