Plain-English Summary
A smaller SD3.5 model designed to run on consumer hardware while preserving good prompt following.
Stable Diffusion 3.5 Medium
A smaller SD3.5 model designed to run on consumer hardware while preserving good prompt following.
Stable Diffusion 3.5
Image
Open weights
Open weights: Yes
API: Yes and self-hosted
Technical Notes
Approximately 2.5B-parameter diffusion-transformer model for customizable text-to-image generation on more accessible GPUs.
Best For
Consumer GPU generation, local creative tools, fine-tuning
Watch Out For
Lower maximum fidelity and composition quality than SD3.5 Large.
Model Specs
- Maker
- Stability AI
- Country
- UK
- Family
- Stable Diffusion 3.5
- Type
- Image
- Release date
- 2024-10-29
- Date confidence
- Exact
- Status
- Open weights
- Context window
- Varies
- Max output
- Not listed
- Parameters
- 2.5B
- Active parameters
- 2.5B
- Architecture
- Multimodal diffusion transformer
- Reasoning
- Not applicable
- Tool calling
- No
- Structured output
- No
- API available
- Yes and self-hosted
- Open weights
- Yes
- Self-hostable
- Yes
- License
- Stability AI Community License
- Inputs
- Text, image
- Outputs
- Image
- Approx. price
- See provider pricing
- How to access
- Provider website
- Model / API ID
- stable-diffusion-3.5-medium
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Fine-tuning, LoRAs, self-hosting
- Completeness
- 90 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source