Stable Diffusion, every mention

23 scenes, the whole family · ← back to Stable Diffusion

mentionsepisodesper episode tap a year for its mentions

Verbatim, from the transcripts: the passages where Stable Diffusion comes up

loading…

▶ 1:13 Elad Gil Early staple diffusion wave where those models came out and people are like,

▶ 19:15 Thomas Dohmke While at the extreme other end of the spectrum with web development and AI, you're going to be able, and you're, we're already there, like, you know, just look at a, a ten-year-old, give them, you know, um, a tool like, like Copilot or,…
▶ 40:47 Thomas Dohmke Um, you know, we saw that with image models like Stable Diffusion and now there's the Flux model from, uh, a startup actually not too far from, from my home base in Germany and in the Black Forest and Freiburg, uh, Black Forest Labs is…

▶ 3:13 Erik Bernhardsson And the, the, the real killer app, as it turns out, like we started this company pre-gen AI, but as it turns out that the, the main thing that really started driving all the traction was, was when stable diffusion came out.
▶ 12:41 Elad Gil You, you mentioned earlier that, um, one of the things that really caused early adoption of modal was, um, stable diffusion and sort of these, these open source models around image gen.

▶ 1:45 Elad Gil In terms of the ability to create other types of content, obviously the very first content wave in some sense was simple text-based things on GPT-III like Jasper, and then we had an image gen wave, and that was mid-journey and stable…

▶ 1:09 Suhail Doshi And so, and I think then soon after that, I think Stable to Fusion came out around June or July of that same year.
▶ 1:15 Suhail Doshi And I got early access, maybe a couple of weeks access to early to, uh, SD one four.
▶ 3:10 Suhail Doshi And so at the time with, with, um, images, there was just sort of, I think there was Midjourney, there was OpenAI doing some DALI stuff, and then you saw sort of stable diffusion.
▶ 5:21 Suhail Doshi And a lot of people in this space just take stable diffusion and fine tune it or do other approaches like that.
▶ 6:37 Suhail Doshi Stable Diffusion XL's architecture, which is like a unit, right? 3 times in the scene
▶ 17:18 Suhail Doshi There's this thing called like, I think MMDIT, which I think Stable Division III is supposed to be based on by that research team at Stability AI.

▶ 12:48 Tuhin Srivastava They use Stable Diffusion image models to generate images.

▶ 12:27 Sarah Guo I remember, um, like a year ago with things like mid journey or with stable diffusion, like, you know, progressively better diffusion models in general, there was a vein of like, oh, like,
▶ 36:27 Elad Gil You know, if you basically look at the last year, and it's only been a year since ChatGPD came out, right, and it's been a year and change since Mid Journey and Stable Diffusion came out,

▶ 10:13 Elad Gil Interesting shift has been happening because basically if you go back a year and a half, Mid Journey launched, Staple Diffusion came out, DALI-II came out, and there's a whole wave of people saying that they were going to go build on… 3 times in the scene

▶ 6:31 Kawal Gandhi I'm really excited to see like LAMA, stable diffusion, other variations coming onto the platform.

▶ 15:59 Clem Delangue Um, and then on other domains, like for example, you know, text classification, uh, information extraction, or you'll be like image generation with stable diffusion and stuff like that.

▶ 6:56 Emad Mostaque And the models when they come out are a new type of programming primitive infrastructure, because you can have a stable diffusion that's two gigabytes that deterministically converts a string into an image.
▶ 12:31 Emad Mostaque So with stable diffusion, for example, we optimized it for a 24 gigabyte VRAM GPU.
▶ 25:23 Emad Mostaque We are working with multiple movie studios at the moment who are saving millions of dollars, just implementing stable diffusion by itself, let alone these other technologies. 2 times in the scene
▶ 27:58 Sarah Guo I think you've said that the core training run for the original stable diffusion was 150,000 a 100 hours, which is like not that huge in the grander scheme of things. 2 times in the scene

Turn any episode into a week of clips.