Image
GA
Open weights
Stable Diffusion 3.5 Large
A mature SD ecosystem with clear licensing, run locally
Release2024-10
Context1K
Max output—
Parameters8B
Local / VRAM12–24 GB
LicenseOpen weights
Overview
Stable Diffusion 3.5 Large is a model in the Image category from Stability AI. It was released in 2024-10. The vendor is based in UK. Best for: A mature SD ecosystem with clear licensing, run locally.
Hardware: FP16 ~16 GB; 12 GB with Q8/offload.
Strengths
- 8B MMDiT
- ComfyUI support
- Negative prompts
Limits
- Not as current as FLUX.2/Qwen-Image
- License required above $1M revenue
Specifications
| Vendor | Stability AI |
|---|---|
| Category | Image |
| Release | 2024-10 |
| Status | GA |
| License | Open weights |
| License name | Stability Community License |
| Commercial use | Conditional |
| Parameters | 8B |
| Architecture | MMDiT |
| Context window | 1K |
| Input modalities | Text, Image |
| Output modalities | Image |
| Prompt types | Text, Image, Negative |
| Negative prompts | Yes |
| Max outputs | 1-8 |
| Resolution | 1MP (1024x1024) |
| Local deployment | Yes |
| VRAM | 12–24 GB |
| System RAM | 16–32 GB |
| API access | Yes |
| Pricing | Free weights (annual revenue <$1M) |
| Confidence | High |
| Vendor country | UK |
| Vendor type | Open lab |
Pricing & hardware
Free weights (annual revenue <$1M)
This model can be self-hosted or run locally. Approximate GPU memory: 12–24 GB. Approximate system RAM: 16–32 GB. API access is available.