datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
svg-benchmark
Rapidata Static SVG Generation Benchmark
Built by Rapidata.
This dataset contains 1,918,367 human responses, collected with the
Rapidata Python SDK, comparing how well 42 frontier LLMs generate
static SVGs from text prompts. Each row is a head-to-head comparison between two models' renders of
the same prompt, scored by human annotators on one of three questions (Preference, Coherence, Alignment).
The SVGs are produced as raw <svg> markup by the models, rasterized to 768×768 PNGs… See the full description on the dataset page: https://huggingface.co/datasets/Rapidata/svg-benchmark.svg-stack
Dataset Card for svg-stack
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-stack")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-stack.svg-stack-filtered
Dataset Card for svg-stack-filtered
This is an attempt to replicate the dataset used for SFT in the paper
Rendering-Aware Reinforcement Learning for Vector Graphics Generation
Processed:
Optimized with svgo precision=2
Rasterized with cairosvg[^cairo]
[^cairo] cairosvg doesn't implement all svg features, but matches how the original paper
Filtered based on some heuristics:
Removed any svg that couldn't be rendered with cairosvg (~30%)
Removed solid-color images
Removed some… See the full description on the dataset page: https://huggingface.co/datasets/darknoon/svg-stack-filtered.svg-icons
Dataset Card for svg-icons
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-icons")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-icons.svg-benchmark
Rapidata Static SVG Generation Benchmark
Built by Rapidata.
This dataset contains 1,355,161 human responses, collected with the
Rapidata Python SDK, comparing how well 30 frontier LLMs generate
static SVGs from text prompts. Each row is a head-to-head comparison between two models' renders of
the same prompt, scored by human annotators on one of three questions (Preference, Coherence, Alignment).
The SVGs are produced as raw <svg> markup by the models, rasterized to 768×768 PNGs… See the full description on the dataset page: https://huggingface.co/datasets/FForty7/svg-benchmark.svg-fonts
Dataset Card for svg-fonts
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-fonts")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-fonts.svg-stack-annotatedsvg-stack-simple
Dataset Card for SVG-Stack (Simple)
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-stack-simple.svg-icons-simple
Dataset Card for svg-icons-simple
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-icons-simple.wikipedia-svg
Wikipedia SVG images dataset
This dataset contains over 1.5M (one million and a half) SVG (vectorial) images scraped from Wikipedia, together with their metadata.
Fields description
url - The web address where the image can be found.
description - A brief explanation or summary of the image's content.
source - The original location or document where the image is published or cited.
author - The creator or originator of the image.
date - The date when the image… See the full description on the dataset page: https://huggingface.co/datasets/fereab/wikipedia-svg.svg-fonts-simple
Dataset Card for svg-fonts-simple
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-fonts-simple.svg-stack-labeled
Svg Stack - Labeled
This dataset consists of the central storage for all datasets related to the SVG Stack dataset. I found it to be lovely, detailed, and of decent to extremely good quality upon observing many different icons and logos during the labeling process.
This is the central dataset, and is currently UNDER CONSTRUCTION. Use with caution, and be aware that the format HAS NOT been frozen. I will make a post announcing when I freeze this dataset, as that will also be the… See the full description on the dataset page: https://huggingface.co/datasets/MrOvkill/svg-stack-labeled.svg-animal-illustrations
SVG Pet Illustrations Dataset
A dataset of 1,416 prompt-SVG pairs for training text-to-SVG generation models, with a focus on cute animal illustrations.
Dataset Description
This dataset contains text prompts paired with their corresponding SVG code, designed for training models to generate vector graphics from natural language descriptions.
Dataset Statistics
Total examples: 1,416
Format: JSONL (JSON Lines)
Fields: prompt, svg
Content
The dataset… See the full description on the dataset page: https://huggingface.co/datasets/yoavf/svg-animal-illustrations.svg-diagrams
Dataset Card for svg-diagrams
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-diagrams")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-diagrams.FIGR-SVG-png-caption
FIGR-SVG-png-caption Dataset
Overview
The FIGR-SVG-png-caption dataset is an extension of the FIGR-8 and FIGR-8-SVG datasets, introduced in the FIGR-8 research paper and its subsequent SVG version (FIGR-8-SVG GitHub repository). This dataset has been enhanced with captions generated by Large Language Models (LLMs) as used in the IconShop project (IconShop paper). Originally in SVG format, the images have been converted to 200 x 200 PNG files for this dataset. This… See the full description on the dataset page: https://huggingface.co/datasets/yxxshin/FIGR-SVG-png-caption.FIGR-SVG
Dataset Card for FIGR-SVG-v2
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/FIGR-SVG-v2")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/FIGR-SVG.starvector-svg-fonts-simple
Dataset Card for svg-fonts-simple
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/mrfakename/starvector-svg-fonts-simple.svg-emoji
Dataset Card for svg-emoji
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-emoji")… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-emoji.Semantic-SVG-Benchmark
Semantic SVG Benchmark
A benchmark of 203 SVG files annotated with human-written semantic object-decomposition trees:
every rendered shape (<path>, <rect>, <circle>, …) in each SVG is assigned to a named semantic
object (e.g. judge, gavel), and objects may be further decomposed into parts
(e.g. Bamboo planter → pot, bamboo). It is the evaluation benchmark of
Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing
(EMNLP 2026). The annotations are ours; the SVGs… See the full description on the dataset page: https://huggingface.co/datasets/KU-MIIL/Semantic-SVG-Benchmark.svg
Synthetic Visual Genome (SVG) Datasets
This repository contains two complementary datasets for VLMs on scene graph understanding and dense visual relationships.
🔗 Related Resources
Training Code: SyntheticVG Repository
Model Checkpoints: ROBIN-3b
Paper: arxiv
Demo: Coming soon
🚀 Quick Start
# For SVG-Relations dataset
import pandas as pd
df =… See the full description on the dataset page: https://huggingface.co/datasets/jamepark3922/svg.svg-emoji-simple
Dataset Card for svg-emoji-simple
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/starvector/svg-emoji-simple.synthetic_wild_svgsvgen_500k_rasterized_jsonified_uuided
SVGEN RJU - SVGEN 500k: Rasterized, JSONified, UUID'ed
I have selected every svg image from svgen that would rasterize under cairosvg, which is significantly less than a 1% failure rate. Under development.
Reasoning
This is the 1st of many SVG datasets I am collecting, extracting, and rasterizing in an attempt to produce a meaningfully helpful spatial reasoning and vertex manipulation model.
Usage
The rasterized images are in PNG format, as bytes. They may be… See the full description on the dataset page: https://huggingface.co/datasets/MrOvkill/svgen_500k_rasterized_jsonified_uuided.SVGenius
SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation
We introduce SVGenius, the first large-scale complexity-stratified benchmark accessing (M)LLMs capabilities across three progressive dimensions: Understanding (perceptua and semantic QA), Editing (bug fixing, code optimization, style editing), and Generation(text-to-SVG, image-to-SVG, style transfer). Built on real-world data from 24 application domains with systematic complexity… See the full description on the dataset page: https://huggingface.co/datasets/xiaoooobai/SVGenius.sketchy-svgs
Dataset Card for "sketchy-svgs"
More Information needed
svg-rendered
SVG to PNG Rendered Dataset
Dataset Summary
This dataset is a processed version of the svgen-500k-instruct dataset, where SVG images have been converted to PNG format for easier consumption in computer vision and machine learning pipelines. Each successfully converted image maintains the original SVG's visual representation while providing a standardized raster format.
Data Fields
png_processed: Boolean flag indicating whether the conversion was successful… See the full description on the dataset page: https://huggingface.co/datasets/thesantatitan/svg-rendered.FIGR-SVG-png-512-caption
FIGR-SVG-png-512-caption Dataset
Overview
The FIGR-SVG-png-512-caption dataset is an extension of the FIGR-8 and FIGR-8-SVG datasets, introduced in the FIGR-8 research paper and its subsequent SVG version (FIGR-8-SVG GitHub repository). This dataset has been enhanced with captions generated by Large Language Models (LLMs) as used in the IconShop project (IconShop paper). Originally in SVG format, the images have been converted to 512 x 512 PNG files for this dataset. This… See the full description on the dataset page: https://huggingface.co/datasets/yxxshin/FIGR-SVG-png-512-caption.starvector-svg-fonts
Dataset Card for svg-fonts
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset = load_dataset("starvector/svg-fonts")… See the full description on the dataset page: https://huggingface.co/datasets/mrfakename/starvector-svg-fonts.simple-shapes-svgThe goal of this dataset is to measure and improve the ability of VLMs to see accurately in spatial dimensions.
I've tried to ensure that all of the examples are not too hard
have sufficient contrast between foreground and background
shapes are not clipped or ambiguous
solid background
canvas is square 512x512
Initially, I've kept the "canvas" that they're working with 512x512 points, but you can learn more by experimenting with the dimensions as well.
starvector-svg-stack-simple
Dataset Card for SVG-Stack (Simple)
Dataset Description
This dataset contains SVG code examples for training and evaluating SVG models for image vectorization.
Dataset Structure
Features
The dataset contains the following fields:
Field Name
Description
Filename
Unique ID for each SVG
Svg
SVG code
Usage
from datasets import load_dataset
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/mrfakename/starvector-svg-stack-simple.