Generative Art Timeline · Chapter 10: On-Chain Era (2020s)

2021

Latent Diffusion Models Introduced

Invention · Europe · Top Moment

Astronaut in a white spacesuit riding a chestnut horse across a dusty lunar landscape
VulcanSphere, Astronaut Riding a Horse (SDXL), 2023. Generated with Stable Diffusion XL in NightCafe Studio from the prompt "a photograph of an astronaut riding a horse". Wikimedia Commons (public domain / CC0).

Unlike DALL-E, which generated images autoregressively from discrete image tokens, Latent Diffusion Models (LDMs) operated in latent space and were the step that enabled text-to-image technology to go mainstream. Instead of working directly with pixels like Diffusion Models, LDMs first compress the image into a simpler, lower-dimensional space (called the latent space) using a pre-trained autoencoder. This compressed version of the image retains the essential information but is much less detailed. The diffusion model then works in this simpler space, which makes the process faster and less resource-intensive. Once the model has generated the image in the latent space, it’s converted back into a full, high-resolution image. Once this capability was introduced, text-to-image modeling became much less expensive and therefore widely available—and impactful.

Text by Peter Bauman. Generative Art Timeline, Le Random.

Sources

Movements

Machine Learning Art

Filed under

Technologies
artificial intelligence, autoencoder, DALL-E, diffusion models, generative AI, latent diffusion models (LDMs), latent space, text-to-image

Related moments

Echoes across time

Related moments

Le Random editorials

Explore all Le Random editorials and artist interviews

Le Random podcast