BREAKING: I'm thrilled to announce that @clipdropapp has been acquired by @StabilityAI 💥 It's been a *wild* journey so far: from the ARCutPaste demo to @ycombinator W21, our seed led by @nathanbenaich and scaling to millions of users 🚀 Next chapter → stability.ai/blog/stability-…

Didn't have much time to dig in during the holidays, but TAESD & CoreML compilation gets it to ~2.5fps. Still ~260ms for 1 forward pass of the 1.8k unet ops on M1. Anyway, that's it for now - Happy new year y'all! 🥂

Vanilla SD turbo runs at ~1.5fps on my M1 Max with the diffusers lib. I wonder how close to realtime it can get? Maybe CoreML compilation, 6bits palettisation, TAESD and a few more tricks like RCFG from StreamDiffusion can push it to ~20fps?

Raising a non-AI pre-seed/seed in 2024

$20/m!

Took me a while to setup all my eth stuff again, but had to do it for this incredible piece: @ciphrd × @zzznah × @fx_hash_ × some kind of mezmerizing neural cellular automata (?)

"‘I have no regrets. My life is squares, triangles, lines" -- Vera Molnár

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform @Clipdropapp, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more: stability.ai/news/stability-…

Are you fast or beautiful? SDXL: Beautiful LCM: Fast SDXL Turbo: Yes

Today, we are releasing Stable Video Diffusion, our first foundation model for generative AI video based on the image model, @StableDiffusion. As part of this research preview, the code, weights, and research paper are now available. Additionally, today you can sign up for our waitlist to access a new upcoming web experience featuring a Text-To-Video interface. To access the model & sign up for our waitlist, visit our website here: bit.ly/40OM53M

I created the Google Cardboard almost ten years ago but quit VR soon after to focus on Machine Learning. VR is based on a ridiculous misunderstanding. Let me explain why in this thread.

Today, @StabilityAI releases Stable Audio 🔈🎶 A latent-diffusion model trained on an exclusive dataset of 800k high-quality samples. It can generate any kind of long-form music tracks and sound effects! Congrats @ednewtonrex and the whole Stable Audio team 🙌

👋 I'm excited to unveil @airstreet’s second fund of $121,212,121 as we accelerate our mission to back ambitious AI-first companies in North America and Europe! 🧵 My reflections on the journey, opportunity and what this means for our founders and community:

Don't blink! Stability. AI just released to Huggingface two models comparible to ChatGPT! 🔥 - FreeWilly 1 (LLaMA 1): huggingface.co/stabilityai/F… - FreeWilly 2 (LLaMA 2): huggingface.co/stabilityai/F…

🚀 @StabilityAI is introducing Stable Doodle! Transform your doodle into images in seconds. 🖌️Try it now on clipdrop.co/stable-doodle #StableDoodle #StableDiffusion

SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis paper page: huggingface.co/papers/2307.0… present SDXL, a latent diffusion model for text-to-image synthesis. Compared to previous versions of Stable Diffusion, SDXL leverages a three times larger UNet backbone: The increase of model parameters is mainly due to more attention blocks and a larger cross-attention context as SDXL uses a second text encoder. We design multiple novel conditioning schemes and train SDXL on multiple aspect ratios. We also introduce a refinement model which is used to improve the visual fidelity of samples generated by SDXL using a post-hoc image-to-image technique. We demonstrate that SDXL shows drastically improved performance compared the previous versions of Stable Diffusion and achieves results competitive with those of black-box state-of-the-art image generators.

Introducing the latest release from Stability AI: Breaking barriers with #SDXL 0.9! SDXL 0.9 produces massively improved text-to-image and composition detail over the beta release and provides a leap in use cases for generative AI imagery. #StabilityAI Unleash your creativity today! → bit.ly/3Xn12bI

Ok, this one's a banger 🧨 Huge leap forward in quality & expressivity

I have to admit, I was not expecting to be so pleased with Clipdrops Uncrop. It easily matches and in some cases outstrips Adobe's Gen Fill in ease of use and being able to do much larger swaths. Plus...it's free... More tests in 🧵

It's great to see all the positive responses to #Uncrop stability.ai/blog/clipdrop-l… It started as onboarding tests with #SDXL and @StabilityAI 's incredible GPU HPC. The model is not perfect yet (it was trained a month ago), but we learned a lot and the next iteration should be 🧨💥