[Community Notebook] Add Super Resolution with Stable Diffusion X4 Upscaler

[jeannineshiu]

This notebook demonstrates image super resolution using the
stabilityai/stable-diffusion-x4-upscaler model.

• Shows how to load and run the SR pipeline
• Includes visual comparison of LR vs SR results
• Runs on free Colab T4 GPU
• Complements the existing latent_diffusion_upscaler.ipynb
  with a more recent SD X4 approach

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[pcuenca]

Looks good to me, what do you think @sayakpaul?

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

sayakpaul commented on 2026-03-27T10:08:25Z
----------------------------------------------------------------

For some reason "#️️Image Super Resolution with HuggingFace Diffusers" is not rendering properly.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

sayakpaul commented on 2026-03-27T10:08:25Z
----------------------------------------------------------------

Line #13.    )

(nit): we could do the device placement here instead of the next line.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

sayakpaul commented on 2026-03-27T10:08:26Z
----------------------------------------------------------------

Line #2.    prompt = "a high quality, sharp photograph"

Would it make sense to expand on this section to guide the users about the recommended hyperparameters? Like what happens if they use a different prompt?

[jeannineshiu]

Hi @sayakpaul, I've addressed all the review comments — fixed the title rendering, merged .to("cuda") into one line, and added a prompt & hyperparameter guidance section. Please take a look when you have a chance!