Model Overview

Note: You need to request the model checkpoint and license from Stability AI

Request the model checkpoint from Stability AI

Description:

Stable Diffusion 3 Medium is Stability AI's text-to-image model that features greatly improved performance in image quality, typography, complex prompt understanding, and resource-efficiency. Compared to previous versions of Stable Diffusion versions, Stable Diffusion 3 leverages a novel Multimodal Diffusion Transformer (MMDiT) architecture that combines a diffusion transformer architecture and flow matching. We have optimized the model's speed using NVIDIA's TensorRT (TRT), resulting in significant acceleration with no quality loss.

Developed by: Stability AI
Model type: Generative text-to-image model

Model Card

Stable Diffusion 3 Model Card

Terms of use

By accessing this model, you are agreeing to the Stable Diffusion 3 terms and conditions of the license, acceptable use policy and stability.ai privacy policy

Third-Party Community Consideration:

This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party’s requirements for this application and use case; see Stability-AI's Stable Diffusion 3 Medium Model Card.

References(s):

• Scaling Rectified Flow Transformers for High-Resolution Image Synthesis paper
• Stability-AI's Stable Diffusion 3 Medium Model Card webpage

Model Architecture:

Architecture Type: Transformer and Convolutional Neural Network (CNN)
Network Architecture: Multimodal Diffusion Transformer
Model Version: Stable Diffusion 3 Medium

Input:

Input Format: Text
Input Parameters: steps, classifier-free guidance scale, output image aspect ratio, negative prompt

Output:

Output Format: Red, Green, Blue (RGB) JPEG Image
Output Size: 1024x1024, 768x1344, 1344x768, 1344x768, 1344x768, 1344x768, 1216x832
Output Parameters: 2D

Software Integration:

Supported Hardware Platform(s): Hopper, Ampere/Turing
Supported Operating System(s): Linux

Inference:

Engine: Triton
Test Hardware: Other

Ethical Considerations

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. Please report security vulnerabilities or NVIDIA AI Concerns here.