Mage: Create Anything. Yes Anything

[hunkins]

Hi all, creator here! Should have put the "Show Hackernews:" entry on the post. Would love feedback and thoughts on next features! Will be posting progress on Twitter (@ghunkins) if interested in updates.

[gault8121]

The loading bar is extremely subtle and hard to see. I missed the tooltip message in the corner. It would be helpful if there was more clear visual feedback that the query is being processed or if the tooltip stayed open until the query was finished.

[ok_dad]

Everyone else is making tool for all of this hires stuff, you should try out an icon generation tool or a small pixel art tool.

Or maybe something that can turn a persons photo into a character from their favorite game style; so a tool which can convert a photo to, for example, a Skyrim-styled character.

I’m not sure if any of that is possible.

[hunkins]

Twitter is actually @greghunkins. Tired.

[WantonQuantum]

Presumably this is Stable Diffusion on the back end. Simple and works well. Any plans for more features?

[hunkins]

Creator here. Yes, hacked this together over the past two days. Community is going wild, and planning to integrate everything sensible.

Next will be: (1) img2img (image-to-image): combine prompt with a given image (i.e. turn your cat into a renaissance painting by Leonardo Da Vinci) (2) inpainting: use text to fill-in an area of a photo (3) outpainting: expand a photo beyond its bounds

Would love thoughts & feedback though.

[qualudeheart]

Eagerly awaiting the proliferation of stable diffusion front ends.

[enneff]

Saying “create anything” is extreme hyperbole.

[hunkins]

Specifically images, so admittedly yes. Generating the right prompt is a tricky science, but if what you're trying to generate contains elements found on the internet, you're probably in luck.

https://lexica.art is a great place for prompt inspiration.

[eimrine]

I am helplessly struggling to generate a man with three hands. All I have is one hand or 2.5 but not 3.

[hunkins]

My best attempt: https://twitter.com/greghunkins/status/1571025487362953218?s...

[hunkins]

Unfortunately, these models don't do great with counting.. (yet!) Might want to try a higher guidance scale if you need the image closer to your prompt.

[hunkins]

Just fixed a speed issue with the API. Pesky async snuck into the code. Should be much snappier now!