Adding images to vibe editor

[aliabd]

Can upload images with the prompt. If images are uploaded, editor will switch to using a vision language model instead of gpt-oss. Still playing around with the different models and the prompt to find the right setup - but its not bad as is

upload-images.mp4

Using zai-org/GLM-4.5V, best so far, though still gets some things wrong.

[gradio-pr-bot]

🪼 branch checks and previews

•	Name	Status	URL
	Spaces	ready!	Spaces preview
	Website	ready!	Website preview
	Storybook	ready!	Storybook preview
🦄	Changes	detected!	Details

---

Install Gradio from this PR

pip install https://gradio-pypi-previews.s3.amazonaws.com/1b697bcc1bce300a2afb9c201820196b47f56b41/gradio-5.46.0-py3-none-any.whl

Install Gradio Python Client from this PR

pip install "gradio-client @ git+https://github.com/gradio-app/gradio@1b697bcc1bce300a2afb9c201820196b47f56b41#subdirectory=client/python"

Install Gradio JS Client from this PR

npm install https://gradio-npm-previews.s3.amazonaws.com/1b697bcc1bce300a2afb9c201820196b47f56b41/gradio-client-1.19.0.tgz

[gradio-pr-bot]

🦄 change detected

This Pull Request includes changes to the following packages.

Package	Version
@gradio/vibeeditor	minor
gradio	minor

---

• Adding images to vibe editor

---

✅ Changeset approved

• Maintainers can remove approval by unchecking this checkbox.

Something isn't right?

• Maintainers can change the version label to modify the version bump.
• If the bot has failed to detect any changes, or if this pull request needs to update multiple packages to different versions or requires a more comprehensive changelog entry, maintainers can update the changelog file directly.

[freddyaboulton]

I couldn't get the GLM model to work 😢 It failed on every prompt, even a simple "make a demo that looks like this" and the demo is a calculator.

It also failed when I uploaded a screenshot of another gradio demo and asked it to update the demo to look like the screenshot. I think we need to update the prompt to make better use of theming I think most models would have a hard time of that tbh.

@AK391 do you have thoughts on how we can improve this?