Stable Diffusion 3 Models by Stability AI
Last refreshed 2026-05-09. Next refresh: weekly.
Capabilities
About
Stability AI's third-generation Stable Diffusion family, built on Multimodal Diffusion Transformer (MMDiT) architecture. The SD 3.5 series includes Medium (2.5B), Large (8.1B), and Large Turbo variants, offering a balance of consumer-hardware accessibility and professional-grade quality. Open-source under the Stability AI Community License.
Decision facts
- Best fit
- image generationvision and multimodal work
- Capability starting point
- Stable Diffusion 3.5 Medium with multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- Stable Diffusion
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs image generation, 2.5B parameters, and multimodal inputs.
Use when the workload needs image generation, 8.1B parameters, and multimodal inputs.
Use when the workload needs image generation, 8.1B parameters, and multimodal inputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Stable Diffusion 3.5 Medium | Use when the workload needs image generation, 2.5B parameters, and multimodal inputs. | 2024-10 | image generation2.5B parametersmultimodal inputs | Current |
| Stable Diffusion 3.5 Large | Use when the workload needs image generation, 8.1B parameters, and multimodal inputs. | 2024-10 | image generation8.1B parametersmultimodal inputs | Current |
| Stable Diffusion 3.5 Large Turbo | Use when the workload needs image generation, 8.1B parameters, and multimodal inputs. | 2024-10 | image generation8.1B parametersmultimodal inputs | Current |
Release Timeline
1 release groupSpecifications(3 models)
| Model | Released | Parameters | Multimodal |
|---|---|---|---|
| Stable Diffusion 3.5 Medium | 2024-10 | 2.5B | Yes |
| Stable Diffusion 3.5 Large | 2024-10 | 8.1B | Yes |
| Stable Diffusion 3.5 Large Turbo | 2024-10 | 8.1B | Yes |
