models/Stable Diffusion 3 (SD3) - SD3 Medium

Stable Diffusion 3 (SD3) - SD3 Medium

7/7/2025

9:42:24 AM

Related Keywords & Tags

base model,basemodel,checkpoint,multimodal diffusion transformer,sd 3,sd3,sd3 medium,stability ai,stable diffusion 3,stable diffusion 3 (sd3),text-to-image model,theally

A striped cat running on hind legs holding a large silver fish with a fishmonger chasing in a crowded open-air market street.

A sad-faced female with long dark hair wearing a white bikini top and shorts holding a wooden sign that says 'Free the melons' in a lush jungle background.

A vibrant Azeroth night elf forest landscape with tall trees, lush green grass, a serene reflective river, and dramatic shadows and highlights in 4K HDR quality.

Glowing semi transparent female figure surrounded by blue magical energy waves against a starry deep space background with volumetric lighting.

A vivid volcano eruption with bright molten lava flowing down its rocky slopes into the ocean, emitting thick black smoke against a dark cloudy sky.

A massive fiery space volcano erupting molten lava and smoke over a sprawling urban cityscape under a clear blue sky with scattered clouds.

A radiant angelic figure named Seraphina with large white wings, flowing ethereal robes, standing in a glowing celestial landscape under a bright divine light.

Hyper realistic portrait of a young woman with long dark hair tied in a ponytail, wearing an intricately detailed black lace dress, with light blue eyes and a serene expression against a dramatic blue and purple background.

Movie poster featuring a mechanical human-tank cyborg girl with short blonde hair on green grass under a blue sky with large orange text reading 'I Can Go Everywhere Now'.

Portrait of an armored war turtle cyborg with detailed blue and yellow mechanical armor, rendered in a renaissance painting style with high resolution and intricate textures.

Portrait of a young woman with glasses, adorned with white flowers in her hair, illuminated by soft ethereal lighting in a surrealistic style influenced by Japanese mythology.

Close-up portrait of a young woman with gold dust on her face, glossy red lips, warm glowing light, and embers floating around her in a dark background.

Tips

Please read the Quickstart Guide for Stable Diffusion 3 at https://education.civitai.com/quickstart-guide-to-stable-diffusion-3/

Refer to the pre-release SD3 Overview for detailed information about the model: https://education.civitai.com/stable-diffusion-3-pre-release-overview/

For more technical details, consult the research paper at https://stability.ai/news/stable-diffusion-3-research-paper

Version Highlights

The 2B Parameter "Medium" weights. Note: no Text Encoder included! Read the Quickstart Guide for more information!

Creator Sponsors

Please read SD3 Unbanned: Community Decision on Its Future at Civitai

Stable Diffusion 3 (SD3) 2B "Medium" model weights!

Please note; there are many files associated with SD3. They will all appear on this model card as the uploads are completed.

ℹ️ Read the Quickstart Guide!

Check out our pre-release SD3 Overview for some information about the model, and the SD3 Quickstart Guide to help you get started generating!

There are three .safetensors versions;

sd3_medium.safetensors
sd3_medium_inc_clips.safetensors
sd3_medium_inc_clips_t5xxlfp8.safetensors

Stable Diffusion 3 Medium is a Multimodal Diffusion Transformer (MMDiT) text-to-image model that features greatly improved performance in image quality, typography, complex prompt understanding, and resource-efficiency.

For more technical details, please refer to the Research paper.

Please note: this model is released under the Stability Non-Commercial Research Community License. For a Creator License or an Enterprise License visit Stability.ai for commercial licensing details.

Model Description

Developed by: Stability AI
Model type: MMDiT text-to-image generative model
Model Description: This is a model that can be used to generate images based on text prompts. It is a Multimodal Diffusion Transformer (https://arxiv.org/abs/2403.03206) that uses three fixed, pretrained text encoders (OpenCLIP-ViT/G, CLIP-ViT/L and T5-xxl)

License

Non-commercial Use: Stable Diffusion 3 Medium is released under the Stability AI Non-Commercial Research Community License. The model is free to use for non-commercial purposes such as academic research.
Commercial Use: This model is not available for commercial use without a separate commercial license from Stability. We encourage professional artists, designers, and creators to use our Creator License. Please visit https://stability.ai/license to learn more.

The original Huggingface Repository for SD3 can be found here.