Stable Diffusion 3.5 Released — a Three-Model Lineup of Large, Turbo, and Medium
Stability AI released the latest version of its open-weight model, “Stable Diffusion 3.5.” True to the company’s description of it as its “most powerful model lineup yet,” it offers three variants to choose from depending on the use case, continuing to be provided under a permissive license.
Details
- Model lineup: The high-performance “3.5 Large” and the fast “3.5 Large Turbo” launched on October 22, followed by the lightweight “3.5 Medium” on October 29
- Runtime environment: All models were designed with an emphasis on customizability and efficiency so they can run on consumer-grade hardware
- Improved diversity: Announced improvements in the diversity of people and compositions in generated images
- License: Free to use for both commercial and non-commercial purposes under the Stability AI Community License
- Distribution: Models released on Hugging Face, with inference code published on GitHub
What happened next
With the release of SD3.5, Stability AI completed its transition from the SDXL to the SD3 model line, greatly expanding the options available for open-weight image generation. The company went on to expand into video and audio generation as well.
If your own PC doesn’t have enough GPU power, ConoHa VPS offers GPU server templates for Automatic1111 (the Stable Diffusion web UI), letting you get started right away with no setup required.
PR
