So when using DiffuserPipeline its easy to get deterministic stable diffusion images by passing a seeded torch.generator. But i also want to be able to get deterministic results when making requests to my stable diffusion 2.1 dedicated endpoint. I have looked up and down the docs but couldn’t find any mention of a seed or generator which can be passed as a dedicated endpoint param. Anybody got a clue? Or is there something i got fundamentally wrong, and i actually have to create a pipeline from my dedicated endpoint which i can then pass the generator to?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| How to optimize inference of stable diffusion model when the images generated are of different seed but with same prompt? | 2 | 1411 | February 7, 2024 | |
| Has Anyone Successfully Deployed FLUX on Hugging Face Inference Dedicated Endpoint? | 2 | 347 | September 9, 2024 | |
| Deploy and add height & width params to Stable Diffusion 2.0 Inference Endpoint | 1 | 907 | February 13, 2023 | |
| Stable Diffusion Inpainting Models fail to deploy | 2 | 1052 | January 4, 2023 | |
| Image-to-image Stable Diffusion Inference Endpoint? | 15 | 3214 | November 29, 2022 |