Skip to main content
  1. Home
  2. Computing
  3. News

Digital Trends may earn a commission when you buy through links on our site. Why trust us?

Stable Diffusion aims to fix its problem with generating fingers

By

Future iterations of AI-generated art are set to be more realistic thanks to an upcoming version of Stable Diffusion that specifically tackles the problem of depicting fingers and hands.

According to a recent Bloomberg report, the company Stability AI, which develops the Stable Diffusion AI image generator, has plans to release a new SDXL 0.9 model that will propel the abilities of Stable Diffusion.

A new Stable Diffusion SDXL 0.9 model sample from Stability AI.
StabilityAI

Stability AI shared a blog post on Thursday, which has since been deleted, detailing the specs and launch details of the SDXL 0.9 model. This leaves questions on exactly what Stability AI has planned for the update.

Recommended Videos

However, from what the post revealed, the new model succeeds the Stable Diffusion XL version that was released in April and will focus on improving hand generation and overall “image and composition detail,” according to Bloomberg.

Please enable Javascript to view this content

The blog post includes sample images generated from the same prompts to show the improvement in quality between the Stable Diffusion XL beta and SDXL 0.9, with the brand saying that the new model stands as “a leap in creative use cases for generative AI imagery.” Some of the prompts include aliens, a wolf, and a person holding a coffee cup.

The upcoming SDXL 0.9 update also follows the Midjourney v5 rollout launched in March, which also focused on improving hand generation. Midjourney AI develops its own proprietary models and has a similar issue where earlier models often generated with an incorrect number of digits on hands, anywhere from four to between seven and 10 on human subjects.

PC compatibility for SDXL 0.9 includes a minimum of 16GB of RAM and a GeForce RTX 20 (or higher) graphics card with 8GB of VRAM, in addition to a Windows 11, Windows 10, or Linux operating system. The model is expected to work through Stability AI’s Clipdrop web tool and will also be added to the company’s DreamStudio app. As per the deleted blog post, there will also be an open-source SDXL 1.0 version. It said this version would launch in mid-July, but now this date is uncertain.

Stable Diffusion is also the source code behind many popular AI image generators, including Starry AI and Night Cafe. Once the SDXL 0.9 update becomes available, it will likely benefit the other partner generators as well.

Editors’ Recommendations

Topics
Fionna Agomuoh
Fionna Agomuoh
Computing Writer
Fionna Agomuoh is a Computing Writer at Digital Trends. She covers a range of topics in the computing space, including…
DALL-E 3 could take AI image generation to the next level
DALL-E 2DALL-E 2 Image on OpenAI.

OpenAI might be preparing the next version of its DALL-E AI text-to-image generator with a series of alpha tests that have now been leaked to the public, according to the Decoder.

An anonymous leaker on Discord shared details about his experience, having access to the upcoming OpenAI image model being referred to as DALL-E 3. He first appeared in May, telling the interest-based Discord channel that he was part of an alpha test for OpenAI, trying out a new AI image model. He shared the images he generated at the time.

Read more
5 things AI image generators still struggle with
Dall-E was an early AI leader but hands are not its thing.

AI image generators like Dall-E, Stable Diffusion, Midjourney, and Bing Image Creator produce amazing results, but sometimes they can be incredibly frustrating. With simple prompts containing just a few words, an AI can output impressive images that appear to be professional photographs and convincing art in various styles. However, the same prompt will occasionally create some horrific creature or hilariously flawed rendering.

Negative prompts might help reduce the likelihood of these errors, but complexity can't always save you. Even AI experts struggle with misshapen creatures and unworldly scenes, requiring long hours of refining prompts or touching-up images with a traditional photo editor. For the time being, if you look carefully in the right areas of an image, there's a good chance you'll be able to identify if it was made by a machine.
Hand salad and balls of fingers
AI developers have made progress in the struggle to teach artificial intelligence tools how human hands should look, but there's plenty of room for improvement. If fingers aren't featured prominently, it's easy to miss errors, but it's an ongoing problem.

Read more
An open-source ChatGPT rival was just launched by the Stable Diffusion team
Stability AI's logo appears along with its mascot a stochastic parrot.

The newest challenger to OpenAI's ChatGPT comes from the company that makes the popular AI image generator Stable Diffusion. Known as StableLM, Stability AI developed this open-source chatbot to democratize access to advanced language models.

Stability AI recently announced the alpha version of StableLM, noting that it is a smaller and more efficient solution than most others. StableLM uses just three billion to seven billion parameters, 2% to 4% the size of ChatGPT's 175 billion parameter model.

Read more