Stable Diffusion 3 AI image models are now available with these features

The Stable Diffusion 3 and Stable Diffusion 3 Turbo models were previewed in February. Now, Stability AI is finally making Artificial Intelligence (AI) text-to-image conversion models available to some users. The company will give developers access to the AI models through the Stability AI Developer Platform API. The company is partnering with the Fireworks AI API platform to make the models available to the public. Notably, the company's next-generation AI models have improved comprehension and writing capabilities.

Stability AI announced the limited availability of AI models in their news item and stated: "As shown in the Stable Diffusion 3 research paper, this model equals or exceeds state-of-the-art text-to-image generation systems, such as DALL-E 3 and Midjourney v6, in typography and deadline adherence based on human preference assessments."

The new text-to-image models have two noteworthy improvements. First, it has improved its understanding of the hint text. It now better understands the contextual knowledge contained in the hint and can now generate images closer to the user's desired images. In addition, the spelling capabilities have been improved. This will help when the user wants to generate an image with written words. Earlier, the company emphasized that the AI will be more attentive to the written words and offer better quality results. The overall image quality is also expected to improve.

In the near future, these new artificial intelligence models will also become open source, at least to some extent. The company said that it will soon make the scale model available for self-hosting with Stability AI membership. Stability AI also explained that the new Multimodal Diffusion Transformer (MMDiT) architecture was used to create the model.

In addition to the AI image generators, Stability AI has also invited a limited number of users to participate in the early release of its Stable Assistant, which is currently in beta testing. The AI assistant is powered by Stable Diffusion 3 and Stable LM 2 12B, which adds conversational capabilities. It can generate images from conversations, generate content, and enhance content to match the generated image. It is currently unknown when the company may release the new AI image models to all users.