On Wednesday, Stability AI released weights for Stable Diffusion 3 Medium, an AI image synthesis model that transforms text prompts into AI-generated images. Its arrival has been ridiculed online, however, because it generates images of humans in a way that seems to be a step backwards from other cutting-edge CG models like Midjourney or DALL-E 3. As a result , it can easily produce wild and anatomically incorrect visual abominations.
A thread on Reddit, titled “Is this version supposed to be a joke?” (SD3-2B),” details SD3 Medium’s spectacular failures in rendering humans, particularly human limbs like hands and feet. Another thread, titled “Why is SD3 so bad at generating girls lying on grass?” » shows similar problems, but for entire human bodies.
Hands have traditionally been a challenge for AI image generators due to the lack of good examples in early training datasets, but more recently several image synthesis models appear to have overcome the problem. In that sense, SD3 seems like a huge step backwards for the CG enthusiasts congregating on Reddit, especially compared to recent stability releases like SD XL Turbo in November.
“Not long ago StableDiffusion was competing with Midjourney, now it looks like a joke in comparison. At least our datasets are safe and ethical!” » wrote one Reddit user.
-
An AI-generated image created using Stable Diffusion 3 Medium.
-
An AI-generated image created using Stable Diffusion 3 of a girl lying in the grass.
-
An AI-generated image created using Stable Diffusion 3 that shows mutilated hands.
-
An AI-generated image created using Stable Diffusion 3 of a girl lying in the grass.
-
An AI-generated image created using Stable Diffusion 3 that shows mutilated hands.
-
An AI-generated SD3 Medium image by a Reddit user made with the prompt “woman wearing dress on beach.”
-
An AI-generated SD3 Medium image by a Reddit user made with the prompt “photograph of a person taking a nap in a living room.”
So far, fans of AI images attribute the failure of Stable Diffusion 3’s Anatomy to Stability’s insistence on filtering adult content (often called “NSFW” content) from AI data. SD3 training that teaches the model how to generate images. “Believe it or not, heavily censoring a model also eliminates human anatomy, so… that’s what happened,” one Reddit user wrote in the thread.
Basically, whenever a user focuses on a concept that is not well represented in the AI model’s training dataset, the image synthesis model will confabulate its best interpretation of what the AI model is asking. ‘user. And sometimes it can be downright terrifying.
The release of Stable Diffusion 2.0 in 2022 suffered from similar issues in properly representing humans, and AI researchers quickly discovered that censoring adult content containing nudity can severely hamper a model’s ability to AI to generate accurate human anatomy. At the time, Stability AI had reversed course with SD 2.1 and SD XL, regaining some lost capabilities by heavily filtering NSFW content.
Another problem that can arise when pre-training the model is that sometimes the NSFW filters used to remove images of adults from the dataset are too picky, accidentally removing images that might not be offensive and depriving the model of human representations in certain situations. “(SD3) works fine as long as there are no humans in the image, I think their improved nsfw filter for filtering the training data has decided that any humanoid is nsfw,” one Redditor wrote on the subject.
Using a free online demo of SD3 on Hugging Face, we ran prompts and saw results similar to those reported by others. For example, the prompt “a man showing his hands” returned an image of a man holding two giant hands backwards, although each hand had at least five fingers.
-
An SD3 Medium example we generated with the prompt “A woman lying on the beach”.
-
An SD3 Medium example we generated with the prompt “A man showing his hands”.
Stability AI
-
An SD3 Medium example we generated with the prompt “A woman showing her hands”.
Stability AI
-
An SD3 Medium example we generated with the prompt “muscular barbarian with weapons next to a CRT TV, cinematic, 8K, studio lighting”.
-
An SD3 Medium example we generated with the prompt “A cat in a car holding a can of beer”.
Stability first announced Stable Diffusion 3 in February, and the company plans to make it available in a variety of different model sizes. Today’s release is for the “Medium” version, which is a 2 billion parameter model. In addition to the weights available on Hugging Face, they are also available for experimentation through the company’s stability platform. Weights are available for download and free use under a non-commercial license only.
Shortly after its February announcement, delays in the release of the SD3 model weights inspired rumors that the release was delayed due to technical issues or mismanagement. AI’s stability as a company has fallen into a tailspin recently with the resignation of its founder and CEO, Emad Mostaque, in March and then a series of layoffs. Just before that, three key engineers – Robin Rombach, Andreas Blattmann and Dominik Lorenz – left the company. And its problems go back even further, as news of the company’s dire financial situation has persisted since 2023.
For some Stable Diffusion fans, Stable Diffusion 3 Medium’s failures are a visual manifestation of the company’s mismanagement and a clear sign that things are falling apart. Although the company has not filed for bankruptcy, some users made dark jokes about the possibility after seeing SD3 Medium:
“I guess they can now go bankrupt safely and ethically (sic), after all.”