5 ms·
MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video
- rvz 2mo agoHollywood and the film industry on red alert. Too bad. This is AGI.
- trwhite 2mo agoThe example video just looks like the highly produced art (TV, commercials, games) other people have created. I find it impossible to believe this wasn't trained on other people's work, and there is no protection for it. Terribly sad. A lack of original thinking is coming.
- echelon 2mo ago> A lack of original thinking is coming. As with the arts, 99.9% of people can't use these models to express vision, get attention, or achieve distribution. The game is the same as it has always been. You still need hard work, taste, something important to say, the ability to articulate it, good timing, and luck. Nothing has changed. We can just build faster. What this does enable is for more to be created that caters to a wider variety of interests. It disrupts existing structures of capital allocation, production, and distribution and gives new players a chance to reshape the game. The bar will rise and people will still be running at the same pace on the treadmill. There will be more to see, but less time to see it.
- anonymousd3vil 2mo agoIt is coming up with Seinfeld episodes even with voice overs. Unless it is trained to do so, don't think that would happen. This model is just around the corner from getting banned due to this copyright issue. (which i hope not)
- MSFT_Edging 2mo agoCan't wait for netflix to be topped on unimaginative nonsense meant to have in the background while you scroll your phone.
- toasty228 2mo agoUnlimited slop "content" to fill decomposed brain shaped vessels, the future is bright!
- kyriakos 2mo agoUnfortunately Hollywood been feeding us human generated slop for a long time already. Transition won't be hard.
- jackp96 2mo agoEh, the technical examples are impressive. But normal people really don't like AI. And I don't think that's going to improve if we really start to see job displacement. I see a future here for digital advertising, for design mock-ups, for political propaganda, for robotics model training/world simulation, for fan movies, etc. But the film industry is struggling to get eyeballs and butts in seats already; using a tool that most people inside-and-outside the industry loathe isn't the answer to fixing that. At most, I could see it used for pre-viz/pre-production work in Hollywood.
- SV_BubbleTime 2mo agoI saw the samples people have posted. Immediately deleted LTX2 and WAN folders. Those are completely worthless now. There is some debate on the license for those in the US, UK, EU, plus… no comment other than whew those samples though!
- Maxious 2mo ago"Regions such as the EU, UK, South Korea, and the US are currently developing or enforcing AI-related regulations that may have specific implications for generative video models" You just have to pinkie promise you won't make disney mad and they will send you a licence https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/QA-about-License.md https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/Q...
- SV_BubbleTime 2mo agoI do work animations for fun and internal use only (mostly jokes). I MAY reach out to them.
- razster 2mo agoWorth if you plan to use it for production use. If you use it for personal or as mockup, then what they don't know won't hurt them? ;)
- CamperBob2 2mo agoAnd boy, will this thing make Disney mad. Least-censored model of any type I've ever seen from a major lab.
- fodkodrasz 2mo agoOn one hand: impressive. On the other hands aesthetically it all looks painfully bland and generic.
- SV_BubbleTime 2mo agoThis, today, is the absolute worse this model will ever be. Chill.
- taneq 2mo agoYou say that, but wait till I start trying to prompt it. :P
- _diyar 2mo agoI agree. But is that the model or the prompt? I remember reading a report where people running AI-model instagram account were using insanely long and detailed prompts about the setting, lighting, makeup, pose, disposition, clothing, etc. about their models. Presumably with some reference image of the face / body to remain consistent across images. It‘s not clear to me whether a sufficiently detailed prompt can generate actually interesting video with a natural ”texture” (for lack of a better word).
- razster 2mo agoThat would be the prompt. With the right assistance from Qwen3.5/Ornith I was able to achieve some amazing results. Unfortunately due to their licensing I'm not allowed to use it in the USA, so I had to halt testing.
- iamcoder18 2mo ago[dead]
- gjs278 2mo ago[dead]
- 2mo ago
- Mashimo 2mo ago> 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?
- sheesdev 2mo agoThe mouse render is surprisingly good. Several of those clips stood out to be a pretty big leap in terms of current SOTA models. The only one that looks "off" is the beverage ad video during the can opening clip, it still has that "AI smoothening" effect. Good thing this can be done pretty well using traditional rendering. I feel like for a good while now we'll transition into a process that uses traditional "close-up" rendering/shots + AI generated wide-shots or quick cuts. Exciting, but also troubling. This being open-weights is a massive win for the community though.
- echelon 2mo agoThis is still about a year and a half behind Seedance 2.0/ Seedance 2.5 But it represents a coming price pressure that will face the leading foundation models. Open source will prevent runaway costs. Moreover, it prevents the hair-trigger platform safety checkers from shutting down creative work. Video models are notoriously bad at shutting down a huge number of requests. Creatives will prefer to work on cloud or private GPU clusters. Waiting 10 minutes for a few seconds of 480p is unacceptable. Hobbyists will have fun, but most actual production work is happening in the cloud. Artist's time is worth money, and they like to spin up dozens of concurrent generations at a time to more quickly explore the generation state space and make progress on completing work.
- coder543 2mo agoA 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-video 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 2.5 since H3 is already ahead of where Seedance was 6 months ago.
- echelon 2mo ago> 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 continues to rank Happy Horse as a good model despite the model being absolute garbage. Could it be because they are subject to broad based statistical attacks? It's easy to encode information about the origin of media in either its metadata or output frames. Or maybe there's simply no overlap between creatives and people who click on ELO scores. I've spent thousands upon thousands of dollars generating video. I will stand by the claim that nothing touches Seedance 2.0 / 2.5. It's good that we're getting better open weights. It puts price pressure on the foundation model companies. But these open models are not a substitute for Kling or Seedance yet. Not even close.
- vblanco 2mo agoIm running this on my 4070ti super (16 gb vram), and it takes 10 minutes for a 10-seconds 480p video. but the results are spectacular.
- robbru 2mo ago1 minute for a second of footage, that is awesome! Thanks for sharing.
- mwigdahl 2mo agoIf you wouldn't mind sharing, what's your Comfy workflow for this? I have the same video card setup and would like to give it a shot.
- vblanco 2mo agojust the default one in the link for image-to-video
- Maxious 2mo agoFWIW on a 5080 16GB it takes 3 minutes for 10 seconds 480p video (the mouse video workflow with length changed from 5 seconds to 10 seconds)
- embedding-shape 2mo agoFWIW on a 6000 Pro it takes 68 seconds for 10 seconds 480p video, (cold) same demo workflow as you used. Set "megapixels" to 2.0 (1920x1088) and same video seems to take 5+ minutes, not sure if everything is right/correct at the moment. As far as I can tell, the current ComfyUI nodes don't even do compilation, and I haven't looked into what attention mechanism they're using, but I'm sure with time these durations will come down even more.
- pkroll 2mo agoPeople on reddit have definitely pointed out that sageattention will speed up the renders. And it's literally the first day. Someone will make a distilled 4-8 step LoRA and we're off to the races. Edit: did a couple of 10 second long 864/480 i2v videos on my RTX Pro 6000: sageattention bumps them up 33%, that is to say, 140.89 seconds without sageattention becomes 105.69 with sageattention on (if using the KJ Sageattention node, "allow_compile" doesn't seem to affect it, just "sage_attention" set to "auto" works fine). EasyCache also appears to work, but does affect quality, at least with the default threshold or even down to 0.10. Still, at 0.10 threshold the same render above, with sageattention, is down to 71.33 seconds, so depending on your use case the quality hit might be worth it. Also it seems that with EasyCache the video still matches the un-EasyCached video (with the same seed), so you could use it to do seed hunting.
- embedding-shape 2mo ago> 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 well? Neat with native frame-to-frame generation, but wonder how easy it is to "link" together clips at the intersection, typically the models kind of lose the "momentum" across these stiches, being able to merge things with frame-to-frame between clips might help with this it feels like.
- _diyar 2mo agoAlso begs the question whether this is applicable for high-throughput applications on FPGAs, which are to my novice mind basically LUTs, right? I remember a paper which was posted on HN a few weeks ago where somebody implemented KAN networks in FPGAs, since those can readily be approximated as LUTs.
- yorwba 2mo ago"Modulation weights" here refers to weights used to adjust layer normalization depending on the task (adaLN). General-purpose LLMs generally don't have those in the first place.
- liuliu 2mo agoOne thing similar would be projecting both the head.weight and the final LLM activations into a smaller vector space, since that is basically just cosine similarity ranking step (so that would reduce the head.weight size). But again, it must be tried many times and just not working as well. LLM space is pretty saturated with tricks.
- knicholes 2mo agoWhoah, could this mean we can treat layers like a jpg, where we come up with a formula that estimates the weight values of a layer instead of storing all of the weights?
- satvikpendem 2mo agoI'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 instead the creativity is in the arrangement. I suspect it'll be quite a while until AI gets a good enough aesthetic sense to do this, as even with static HTML websites humans can easily see that it's AI slop.
- ilaksh 2mo agoSeedance 2.5 just came out and it is incredible，significantly better than this for a lot of cases. This one is the latest _free_ video generator. But as far as your composition tools, they are already available.
- unleaded 2mo agoor 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)
- satvikpendem 2mo agoJust like there are still plays even though movies overtook them as the main form of entertainment with acting, so too will AI generation overtake movies with real people. It's simply cheaper and more freeform creatively to use AI so expect that to be the future.
- unleaded 2mo agoThe fundamental difference between AI and all the previous advances like that is that people actually liked and wanted films, synchronized sound, color, CGI, etc. and queued down the block for them. Same thing with electronic music, people enjoy listening to it. Not really the same thing with AI, most of the praise is about the impact it will have on the production process, cost-effectiveness, etc., what actually comes out tends to take the back seat. From what we've seen it's pretty difficult getting people to like the output as well.
- hnlqpx99l9 2mo ago[dead]
- nfnmema 2mo agoAny tutorial for me to learn how to use
- jclarkcom 2mo agoTry this prompt with claude code "download the new minimax h3 model and render a test movie with it"
- storus 2mo agoReference-to-video mode seems like all that was missing to enable completely independent cinematography as right now one couldn't stitch different scenes together properly without altering substantial portions of the scene.
- mihau 2mo agoHas anyone tried running this on Mac device? (e.g. Mac Studio Ultra)
- Havoc 2mo agoFailed with a software error for me
- bhawika_kaushik 2mo ago[flagged]
- torment-nexus 2mo agoThis is an interesting problem. A similar approach that worked for us was to only load the tools the agent actually needs for each task.
- vunderba 2mo agoI 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 turned out overall, and how fast it ran. A 10-second, half-megapixel video gen took only a few minutes which is kinda crazy especially thinking back early WAN days. Video demos: https://imgpb.com/rllwg https://imgpb.com/rllwg
- keepupnow 2mo agoHilarious well done
- nl 2mo agoThat spinning person one would make an interesting benchmark. The model clearly has a strong prior that human heads go upright.
- vunderba 2mo agoOh yeah definitely agree. I think one of the older tests for LTX and WAN text-to-video models used to involve having a person do a cartwheel or really anything that required putting them in a non‑upright orientation.
- emliy-big 2mo ago[flagged]