Model reference · open weights

stable-diffusion-3-medium

Image stabilityai Text→image 2 builds Its own licence terms 37k dl/mo

stable-diffusion-3-medium is an open-weight image model from Stability AI. stable-diffusion-3-medium (BF16) weighs 36.9 GB; the smallest configuration that runs it is RTX 4090 24 GB.

  • Stable Diffusion 3 Medium is a 2.1B parameter Multimodal Diffusion Transformer developed by Stability AI for text-to-image generation.
  • The model uses three fixed, pretrained text encoders and was pre-trained on 1 billion images.
  • It is released under the Stability Non-Commercial Research Community License, which restricts commercial use without a separate license.

Summary of the stabilityai/stable-diffusion-3-medium-diffusers model card, 2026-10-01. The estimate below is for another build of the family.

What it is

Released byStability AI
Released2024-06-12
Parameters2.1B
VRAM36.9 GB for the weights

What it runs on

Memory and cards for stable-diffusion-3-medium (BF16)

36.9 GBweights, file size
15.8 GBbiggest part
5.0 GBworking memory, one 1024² image
537 MBruntime overhead
CardOne imageMemory
RTX 3060 12 GB … RTX 4060 Ti 16 GBdoes not fit
RTX 3090 24 GBtight (encoders offloaded)23.4 GB
RTX 4090 24 GBtight (encoders offloaded)23.4 GB
RTX 5090 32 GBfits (encoders offloaded)31.0 GB
L40S 48 GBtight44.0 GB
A100 80 GBfits78.2 GB
H100 80 GBfits78.1 GB
RTX PRO 6000 Blackwell 96 GBfits93.8 GB
DGX Spark (GB10) 128 GB unifiedfits107 GB
H200 141 GBfits138 GB
B200 180 GBfits176 GB

Builds

Sizes, precisions & builds

BuildParamsPrecisionWeightsSmallest setup
stable-diffusion-3-medium-diffusers ↗ 2.1BFP16 15.5 GBRTX 4060 Ti 16 GB
stable-diffusion-3-medium (above) ↗ —BF16 36.9 GBRTX 4090 24 GB

How it works

How image models work

Text promptwhat to makeText encoderunderstands itDiffusion stepsdenoise to pixelsImagePNG / JPEGA diffusion model starts from noise and denoises it, guided by your prompt, into a finished image.

Running it yourself

Run it on a rented GPU

Rent a machine by the hour. How to run this model is on its model card.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms