On one hand: impressive. On the other hands aesthetically it all looks painfully bland and generic.
This, today, is the absolute worse this model will ever be. Chill.
You say that, but wait till I start trying to prompt it. :P
MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
81–90 of 100 posts
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#82I've said it before and I'll say it again, human directors are still valuable, as they use AI video editing tools to generate the shots they want and put them together in a cohesive way. Previously they might've used film and actors but if they can just prompt the AI (or create workflows as seen with ComfyUI) then they arrange them together just like how an EDM producer doesn't actually play the instruments but inste…
or option 2 we'll keep making productions with film and actors. I get it's easy to feel that it's all over with how good these video models are getting but thinking we'll all be slaves to the slop machine once it gets good enough is pretty pessimistic depressing and IMO unlikely.
(I do think it will get a foothold in the "crap people are ashamed to admit they watch" sector though, which it basically already has)
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#83Earlier quoted context omitted.
> everything we can "know" can be written down Do you know how to ride a bike?
I think that's different though, that's "doing" rather than just "knowing". You can ride a bike without knowing how it works, and obviously vice-versa too. I don't see circuit diagrams as "doing" though, but the soldering part of building circuits definitely is that way though, you can't just read about it and excel first time you pick up an iron, you have to practice and understand it with your body, like bicycling.
[dead]
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#84Earlier quoted context omitted.
A year and a half behind Seedance 2.0? That is a bold claim that needs evidence. According to one user preference leaderboard, MiniMax H3 is already ahead of Seedance 2.0 based on thousands of A/B votes: https://artificialanalysis.ai/video/leaderboard/text-to-vide... I haven't seen any user preference comparisons between Seedance 2.5 and MiniMax H3. As an upper bound, H3 cannot be more than 6 months behind Seedance 2…
> A year and a half behind Seedance 2.0? That is a bold claim that needs evidence. SOTA a year ago was Kling 2.5, and H3 does not look or perform at that level. The Artificial Analysis rankings are whack. They rank Omni first, which is incredulously wrong. Google's models broadly suck, and there they all are - right at the top. Artificial Analysis has notoriously ranked models such as Grok Imagine highly and continue…
On one hand we have your anecdotes, on the other hand, data.
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#85I dug up a few old parody ideas I’d had back in high school and threw them at MiniMax M3 on my RTX. There’s definitely still a lot of jank once you move away from fairly normal scenarios. The moment you start to veer into weirder concepts, things tend to break down a bit especially in the game show where someone is strapped to a wheel and being spun. Still tho, I was actually shocked by how well the text-to-video tur…
That spinning person one would make an interesting benchmark. The model clearly has a strong prior that human heads go upright.
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#86Earlier quoted context omitted.
right before that, there's the part with the person hiking up the dish and 'breathing', and the clouds of water vapor coming out of their mouth don't line up with their breaths devs pls fix
> the clouds of water vapor Really poor. In the last shot, looks more like smoke. The fact the director even included this item in the reel speaks volumes.
reroll
AI psychosis gets everyone, like our human context window drifts the further the session goes on and discernment quality decreases
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#87I've said it before and I'll say it again, human directors are still valuable, as they use AI video editing tools to generate the shots they want and put them together in a cohesive way. Previously they might've used film and actors but if they can just prompt the AI (or create workflows as seen with ComfyUI) then they arrange them together just like how an EDM producer doesn't actually play the instruments but inste…
outdated take
there are sequencing AI that will edit
case in point, the entire SaaS multimedia space has pivoted this year to agentic workflows, as in, no more generating AI for that sensitive audience, but instead automating the human work of editing and compositing of real media
if you so happen to supply generative media it will form a cohesive edit of that too
also website slop is distinct from the AI generated sites that blend in. you only notice the ones that don’t.
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#88> The result gives a total memory footprint reduced by 66%, from 123.6 GB in full precision to 42.5 GB with the smallest models variants. Combining this with our dynamic VRAM offloading enables a next-generation 2K video model to run locally on a GPU like the RTX 3060. Pretty cool. But assuming you have a 16GB 3060, how long would it take to generate a 15 second clip?
I don't believe the 3060 has a 16gb variant, only 12 and 8
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#89Any tutorial for me to learn how to use
Try this prompt with claude code "download the new minimax h3 model and render a test movie with it"
Re: MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
#90> We found that the model's modulation weights (~40% of the total parameters) could be pruned and replaced with a functionally equivalent lookup table, dramatically shrinking the memory footprint with no loss in output quality. Is this a common approach to reducing weights with "no loss in output quality", assuming this is true? Seems almost too simple to work. If this is doable, would this be applicable to LLMs as w…
This isn't the same with lookup tables, but in Explaining Attention with Program Synthesis they were able to replace some attention heads with Python programs: