joanrod/star-vector
StarVector is a foundation model for SVG generation that transforms vectorization into a code generation task. Using a vision-language modeling architecture, StarVector processes both visual and textual inputs to produce high-quality SVG code with remarkable precision.
Built on a StarCoder-derived architecture, StarVector applies semantic understanding to generate compact SVG code using primitives (ellipses, polygons, text) rather than just path curves, addressing vectorization quality issues traditional methods produce. The model operates directly in SVG code space, leveraging vision-language capabilities for both image-to-SVG and text-to-SVG tasks, with 1B and 8B parameter variants available on Hugging Face alongside the SVG-Stack training dataset (2M samples) and SVG-Bench evaluation benchmark spanning 10 datasets.
4,299 stars.
Stars
4,299
Forks
241
Language
Python
License
Apache-2.0
Category
Last pushed
Nov 07, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/joanrod/star-vector"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.