Connect with us


New Turbo Models Now Accessible Through Stability AI Developer Platform API



Stable Diffusion 3, Turbo Models Are Now Available via Stability AI Developer Platform API

Stability AI has officially made its Stable Diffusion 3 and Stable Diffusion 3 Turbo AI text-to-image models available to developers through the Stability AI Developer Platform API. The company has partnered with Fireworks AI to bring these next-generation AI image models to the public. These models come with enhanced text understanding and spelling capabilities.

The announcement of the limited availability of the AI models was made by Stability AI in a post on its newsroom. The company stated that these models are on par with or even surpass state-of-the-art text-to-image generation systems like DALL-E 3 and Midjourney v6 in terms of typography and prompt adherence based on human preference evaluations.

The new text-to-image models have notable upgrades, including improved understanding of prompt text and enhanced spelling capabilities. This means the AI can better comprehend contextual knowledge within prompts and generate images that align more closely with user preferences. Additionally, the AI will pay closer attention to written content, resulting in better image output and overall improved image quality.

Stability AI also plans to open-source these new AI models in the future to some extent. The company will make the model weights available for self-hosting with a Stability AI Membership. The models were developed using a new architecture called the Multimodal Diffusion Transformer (MMDiT).

In addition to the AI image generators, Stability AI has introduced a limited early release of its Stable Assistant, powered by Stable Diffusion 3 and Stable LM 2 12B. This AI assistant, currently in beta, offers conversational capabilities and can generate images from conversations, create content, and enhance content to align with the generated image. The release date for the new AI models to all members has not been disclosed yet.

Click to comment

You must be logged in to post a comment Login

Leave a Reply