News for hackers. Stuff that matters.
Re:NeuralSVG: An Implicit Representation for Text-to-Vector ... by pizza on Wednesday January 08, 2025 @01:29PM (#42637016)
Prompting Claude to make SVGs then dropping them into Inkscape and getting the last ~20% of it to match the picture in my head has been a phenomenal user experience for me. This, too, piques my curiosity..!
Re:NeuralSVG: An Implicit Representation for Text-to-Vector ... by xvfLJfx9 on Wednesday January 08, 2025 @01:33PM (#42637056)
Claide doesn't work at all for me when generating SVGs
Re:NeuralSVG: An Implicit Representation for Text-to-Vector ... by iambateman on Wednesday January 08, 2025 @01:39PM (#42637110)
It depends for me on if there is an existing SVG that exists in its training set.

“Make an SVG of a clock icon” is likely to work. “Make an SVG of a playground swingset with the sun setting” is not.

Re:NeuralSVG: An Implicit Representation for Text-to-Vector ... by varunneal on Wednesday January 08, 2025 @01:45PM (#42637171)
Your prompt verbatim turned out quite well, single-shot.

https://claude.site/artifacts/0f696bf8-399d-42c3-93c0-296493...

Re:NeuralSVG: An Implicit Representation for Text-to-Vector ... by jgalt212 on Wednesday January 08, 2025 @02:07PM (#42637383)
I poked around with NeoSVG a few months back. I was not happy with the results, the computation time, or the cost. That being said, I do hope they've made big progress lately because SVGs work real nice when you have an LLM and human working in tandem (as per the comment above).

https://neosvg.com/generations