Stable Diffusion 3.5 Flash
Stable Diffusion 3.5 Flash is a distilled version of Medium exposed through the current API. The documentation describes four-step generation, compared with the longer sampling process of Medium.
Model design and access
The shared generation endpoint supports text-to-image and image-to-image requests. Image-to-image needs a starting image, strength and the appropriate mode. Flash should be selected explicitly through the model parameter.Stable Diffusion 3.5 API reference
Practical use
Practical guidance. Flash fits applications where quick image proposals matter more than maintaining the exact response of an older checkpoint. Compare it with Medium using representative layouts and subjects. Keep the hosted API contract separate from plans for local inference, since a documented endpoint does not establish downloadable weight access.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | sd3.5-flash |
| Model ancestry | Distilled from Stable Diffusion 3.5 Medium.Stable Diffusion 3.5 API reference |
| Output | One megapixel; 1024 × 1024 by default.Stable Diffusion 3.5 API reference |
| Access | Named hosted API option; no public Flash weights verified.Stable Diffusion 3.5 API reference |
Related models
- Stable Diffusion 3.5 Large Turbo
Stability AI · Stable Diffusion
A distilled Stable Diffusion 3.5 Large image model that generates images in four sampling steps.
Launch date not confirmedAPI · Downloadable - Stable Diffusion 3.5 Medium
Stability AI · Stable Diffusion
Stable Diffusion 3.5 Medium uses the MMDiT-X architecture for image generation.
Launch date not confirmedAPI · Downloadable