> Something being available OSS is very different from a turnkey product solution, not to mention that Tencent's 60 GiB requirement requires a setup with like at least 3-4 GPUs which is quite rare & fairly expensive vs something time-sharing like Sora where you pay a relatively small amount per video. It took two weeks to go from Mochi running on 8xH100s to running on 3090s. I don't think you appreciate the rapidity…
If you've got a 4090 and ComfyUI can you run HunYuan?
There are already Hunyuan fp8 examples running on a 4090 on r/stablediffusion.
Every day that passes I grow fonder of Google's decision to delay or otherwise keep a lot of this under the wraps. The other day I was scrolling down on YouTube shorts and a couple videos invoked an uncanny valley response from me (I think it was a clip of an unrealistically large snake covering some hut) which was somehow fascinating and strange and captivating, and then scrolling down a few more, again I saw someth…
On the other hand, because these tools like this are being made available before output is perfected, you and many others are being trained in AI discernment; being able to detect fake things will be a helpful skill to have for some time: another form of critical thinking. It would be FAR worse if a privately held advanced AI's outputs were unleashed without the population being at least somewhat cautious of everythi…
I think as these things will get bigger and better much faster than we can learn to discern.
Every day that passes I grow fonder of Google's decision to delay or otherwise keep a lot of this under the wraps. The other day I was scrolling down on YouTube shorts and a couple videos invoked an uncanny valley response from me (I think it was a clip of an unrealistically large snake covering some hut) which was somehow fascinating and strange and captivating, and then scrolling down a few more, again I saw someth…
On the other hand, because these tools like this are being made available before output is perfected, you and many others are being trained in AI discernment; being able to detect fake things will be a helpful skill to have for some time: another form of critical thinking. It would be FAR worse if a privately held advanced AI's outputs were unleashed without the population being at least somewhat cautious of everythi…
>you and many others are being trained in AI discernment
HN is a hyper specialized group of people. The average person can not do this and as we've seen devours up misinformation with no second thoughts.
> I’ve worked in CG for many years and despite the online nerd fests that decry CG imagery in films, 99% of those people can’t tell what’s CG or not unless it’s incredibly obvious. I've noticed people assume things are CG that turn out to be practical effects, or 90% practical with just a bit of CG to add detail.
Yep I’ve had that happen many times , where people assume my work is real and the practical is CG. Worse, directors often lie about what’s practical and we’ll have replaced it with CG. So people online will cheer the “practicals” as being better visually, while not knowing what they’re even looking at. I’ve seen interviews with actors even where they talk about how they look in a given shot or have done something, an…
But films without CG are clearly superior and it’s not even in contention.
Lawrence of Arabia or Cleopatra alone have incredible fully live shot special effects which can not be easily replicated with CG and have aged like fine wine, unlike the trash early CG of the 80s and 90s which ruined otherwise great films like the last starfighter
Every day that passes I grow fonder of Google's decision to delay or otherwise keep a lot of this under the wraps. The other day I was scrolling down on YouTube shorts and a couple videos invoked an uncanny valley response from me (I think it was a clip of an unrealistically large snake covering some hut) which was somehow fascinating and strange and captivating, and then scrolling down a few more, again I saw someth…
It saddens me. Innovations in AI 'art' generation (music, audio, photo) have been a net negative to society and are already actively harming the Internet and our media sphere. Like I said in another comment, LLMs are cool and useful, but who in the hell asked for AI art? It's good enough to fool people and break the fragile trust relationship we had with online content, but is also extremely shit and carries no meani…
Go on civil.AI, it’s primarily used for hardcore waifu porn.
On the other hand, because these tools like this are being made available before output is perfected, you and many others are being trained in AI discernment; being able to detect fake things will be a helpful skill to have for some time: another form of critical thinking. It would be FAR worse if a privately held advanced AI's outputs were unleashed without the population being at least somewhat cautious of everythi…
>you and many others are being trained in AI discernment HN is a hyper specialized group of people. The average person can not do this and as we've seen devours up misinformation with no second thoughts.
And yet, OP referred to a thread where the reality of the shorts were being questioned by "average" people. Imagine a world where OpenAI were the first out the gates with this and just started producing their own videos without telling anyone about their technology or letting creators play with it. They'd make loads of money, probably could topple governments... I'm glad these tools are being made generally available versus the alternative.
On the other hand, because these tools like this are being made available before output is perfected, you and many others are being trained in AI discernment; being able to detect fake things will be a helpful skill to have for some time: another form of critical thinking. It would be FAR worse if a privately held advanced AI's outputs were unleashed without the population being at least somewhat cautious of everythi…
I think as these things will get bigger and better much faster than we can learn to discern.
With zero doubt. Faster than we expect. And yet, it's nice that we are learning to distrust what we see before the "real real" stuff comes out.
So is yours! Mine isn't, however. I am a hard-nosed real boy now.
Write something that an LLM could never write. (This is my latest favorite prompt and interview/conversation question)
If you’re not actively publishing at top conferences (I.e. NeurIPS), than this is a trash question and shows the lack of knowledge that many who are now entering the field will have.
Anything that you or others can answer to this which isn’t some stupid “gotcha” puzzle shit (lol it’s video cus LLMs aren’t video models amiright?) will be wrong because of things like structured decoding and the fact that ultra high temperature works with better samplers like min_p.