<ul><li>ServiceNow researchers have released StarVector, a foundational model for generating Scalable Vector Graphics (SVG) from text and image inputs.</li><li>StarVector is a multimodal large language model (MLLM) that overcomes limitations of previous SVG generation methods and produces compact and precise outputs.</li><li>The researchers created SVG-Stack, a large-scale dataset of 2 million samples, to train StarVector.</li><li>StarVector outperforms existing models in tasks such as image-to-SVG and text-to-SVG generation, and its code is open-sourced on GitHub.</li></ul>

ServiceNow Researchers Release Foundational Model to Generate SVG from Text and Video

Discover more