Model Information





Description
The model has been retrained to generate images in the sora generation style, there may be problems with face generation because it was not possible to achieve the desired results without losing the desired style, usually upscale or facedetailer helps
I noticed the best parameters for myself
cfg: 4-5
dpmpp_2m + karras
steps: 25-40
For maximum effect, add: r3a1izm, realistic, realism
Illumireal
166downloads
Download ModelModel Details
- Type
- AI Model
- Task
- text-to-image
- Subtype
- Safetensors / Checkpoint AI Model
- Created
- Updated
- June 7, 2026