[D] Imagic Stable Diffusion training in 11 GB VRAM with diffusers and colab link.

Code: [https://github.com/ShivamShrirao/diffusers/tree/main/examples/imagic](https://github.com/ShivamShrirao/diffusers/tree/main/examples/imagic)

Colab: [https://colab.research.google.com/github/ShivamShrirao/diffusers/blob/main/examples/imagic/Imagic\_Stable\_Diffusion.ipynb](https://colab.research.google.com/github/ShivamShrirao/diffusers/blob/main/examples/imagic/Imagic_Stable_Diffusion.ipynb)

Still need to play around and tune the parameters a bit, may not work as is on every subject. Hopefully everyone can try it out now.

[Input Image](https://preview.redd.it/xp1jkpk3apu91.jpg?width=512&format=pjpg&auto=webp&s=71b178cf6ba698ccf960e25c13d0e4ab86540289)

[A photo of Barack Obama smiling with a big grin.](https://preview.redd.it/nr45lnz3apu91.png?width=512&format=png&auto=webp&s=e6bf6ab64617b4db1f2c32ab0ec7d13f26d5daaf)

colab diffusers diffusion machinelearning stable diffusion training