StarVector (image/text to SVG)
Source: https://github.com/joanrod/star-vector
Structured fields
{
"archived": false,
"revision": {
"ref": "main",
"pinned": false
},
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "Vision-language model family (1B and 8B) that generates SVG code from images or text, with the SVG-Bench benchmark and SVG-Stack dataset; a research route to editable vector illustration and icons rather than raster output.",
"source_type": "repository",
"availability": "available",
"source_owner": "StarVector authors (joanrod)",
"next_check_at": "2026-10-31T00:00:00.000Z",
"known_limitations": [
"Dataset rights unresolved."
],
"upstream_updated_at": "2025-11-07T20:31:24.000Z"
}Evidence claims
- [upstream_claim] README: StarVector is a multimodal VLM for SVG generation, supporting image2SVG and text2SVG by posing generation as code generation; models released on Hugging Face (1B, 8B); SVG-Bench and svg-stack datasets linked. · locator: README intro and news source
- [sourced_fact] HF metadata: starvector-1b-im2svg and starvector-8b-im2svg carry license:apache-2.0; datasets/starvector/svg-stack exposes no license tag. · locator: HF API tags source
- [sourced_fact] Repo metadata: Apache-2.0 code, not archived, last push 2025-11-07. · locator: GitHub API fields source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | Apache-2.0 | permitted | permitted | permitted | unknown | Apache-2.0. |
| model_weights | Apache-2.0 | permitted | permitted | permitted | unknown | HF license tags only; base model licenses (VLM backbone) not checked. |
| dataset | unknown | unknown | unknown | unknown | unknown | svg-stack has no license tag on HF; scraped SVG provenance not assessed. |