FWIW it feels like Google should dominate text/image -> video since they have access to Youtube unfettered. Excited to see what the reception is here.
Everyone has access to YouTube. It’s safe to assume that Sora was trained on it as well.
Veo 2: Our video generation model
31–40 of 342 posts
Re: Veo 2: Our video generation model
#32Random fact: Veo means "I see" in Spanish. Take it on any way you want.
While Video means "I see" in latin
Re: Veo 2: Our video generation model
#33FWIW it feels like Google should dominate text/image -> video since they have access to Youtube unfettered. Excited to see what the reception is here.
Everyone has access to YouTube. It’s safe to assume that Sora was trained on it as well.
All you can eat? Surely they charge a lot for that, at least. And how would you even find all the videos?
Re: Veo 2: Our video generation model
#34FWIW it feels like Google should dominate text/image -> video since they have access to Youtube unfettered. Excited to see what the reception is here.
Everyone has access to YouTube. It’s safe to assume that Sora was trained on it as well.
Does everyone have "legal" access to YouTube.
In theory that should matter to something like Open(Closed)Ai. But who knows.
Re: Veo 2: Our video generation model
#35This looks great, but I'm confused by this part: > Veo sample duration is 8s, VideoGen’s sample duration is 10s, and other models' durations are 5s. We show the full video duration to raters. Could the positive result for Veo 2 mean the raters like longer videos? Why not trim Veo 2's output to 5s for a better controlled test? I'm not surprised this isn't open to the public by Google yet, there's a huge amount of volu…
> I'm not surprised this isn't open to the public by Google yet, Closed models aren't going to matter in the long run. Hunyuan and LTX both run on consumer hardware and produce videos similar in quality to Sora Turbo, yet you can train them and prompt them on anything. They fit into the open source ecosystem which makes building plugins and controls super easy. Video is going to play out in a way that resembles image…
> Hunyuan and LTX both run on consumer hardware
Are there other versions than the official?
> An NVIDIA GPU with CUDA support is required. > Recommended: We recommend using a GPU with 80GB of memory for better generation quality.
https://github.com/Tencent/HunyuanVideo
> I am getting CUDA out of memory on an Nvidia L4 with 24 GB of VRAM, even after using the bfloat16 optimization.
Re: Veo 2: Our video generation model
#36FWIW it feels like Google should dominate text/image -> video since they have access to Youtube unfettered. Excited to see what the reception is here.
They also had a good chunk of the web text indexed, millions of people's email sent every day, Google scholar papers, the massive Google books that digitized most ever published books and even discovered transformers.
Re: Veo 2: Our video generation model
#37The example of a "Renaissance palace chamber" is very historically inaccurate by around a century or two, the generated video looks a lot like a pastiche of Versailles from the Age en Enlightenment instead. I guess that's what you get by training on the internet.
Re: Veo 2: Our video generation model
#38My theory as to why all the bigtech companies are investing so much money in video generation models is simple: they are trying to eliminate the threat of influencers/content creators to their ad revenue. Think about it, almost everyone I know rarely clicks on ads or buys from ads anymore. On the other hand, a lot of people including myself look into buying something advertised implicitly or explicitly by content cre…
I did not think about that angle yet but I have to admit, I agree. I rarely ever even pay attention to the YT ads and kind of just zone out but the recommendations by content creators I usually watch are one of the main sources I keep up with new products and decide what to buy.
Re: Veo 2: Our video generation model
#39Judging by how they've been trying to ram AI into YouTube creators workflows I suppose it's only a matter of time before they try to automate the entire pipeline from idea, to execution, to "engaging" with viewers. It won't be good at doing any of that but when did that ever stop them. https://www.youtube.com/watch?v=26QHXElgrl8 https://x.com/surri01/status/1867433782992879617
They basically already have this: https://workspace.google.com/products/vids/
Last week I started seeing a banner in Google Docs along the lines of "Create a video based on the content of this doc!" with a call to action that brought me to Google Vids.
Re: Veo 2: Our video generation model
#40Judging by how they've been trying to ram AI into YouTube creators workflows I suppose it's only a matter of time before they try to automate the entire pipeline from idea, to execution, to "engaging" with viewers. It won't be good at doing any of that but when did that ever stop them. https://www.youtube.com/watch?v=26QHXElgrl8 https://x.com/surri01/status/1867433782992879617
Who needs viewers anyway? Automate the whole thing. I just see the endgame for the internet is https://en.wikipedia.org/wiki/Dead_Internet_theory