The Checkpoint #12 – This week in AI art including a guide to latent space, mini mini Dall-e, guess the prompt and wonky cats
A traveler’s guide to the latent space – A hugely useful guide to Disco Diffusion from @Ethan_smith_20, covering everything from prompt engineering to deep-dives on the various settings & parameters. Recommended.
Wordalle – Guess the prompt!
Min Dalle – Even more minimal implementation of Dalle-Mini, mini enough to fit on an M1 Pro.
Elucidating the design space of diffusion-based generative models – New state-of-the-art in text-to-image generation. Currently being implemented by Lucidrains as Imagen PyTorch and @RiversHaveWings as k-diffusion.
Training your own unconditional diffusion model (with minimal coding) – Great article from @KaliYuga_ai on how to create your own custom Disco Diffusion model.
Featured notebook – CogView2
Announced a few weeks ago, CogView2 is the largest model publicly available. It's similar to Imagen and Dall-E 2, though the results aren't quite as good. And the code is publicly available, unlike the others.
Thanks for reading!
If you have anything you’d like to be featured or want to get in touch, give me a shout on Twitter or via email. Please also consider supporting me on Patreon so that I can spend more time creating content like this.