OmniSVG

OmniSVG: A Unified Scalable Vector Graphics Generation Model                OmniSVG is the first family of end-to-end multimodal SVG generators that leverage pre-trained Vision-Language Models (VLMs), capable of generating complex and detailed SVGs, from simple icons to intricate anime characters. We also introduce MMSVG-2M, a multimodal dataset with two million richly annotated SVG assets, along with a standardized evaluation protocol for conditional SVG generation tasks. 2. Models Downloading | Model | Download link | Size | Update date | |-----------------------------|-------------------------------|------------|------| | OmniSVG-3B| 🤗 Huggingface | 8.49 GB | 2025-07-21 | 3. Dependencies and Installation The dependencies configured according to the following instructions provide an environment equipped for inference 3.2 Create Conda Environment Create and activate a new conda environment with Python 3.10: System Dependencies Before installing Python packages, you need to install Cairo library which is required by `CairoSVG` in our dependencies: > Note: Installing Cairo system library beforehand helps prevent potential build errors when installing `CairoSVG` via pip. Python Dependencies We have tested our environment with CUDA 12.1. You can install CUDA 12.1 by following the CUDA Toolkit installation guide. | | GPU Memory Usage | Time per 256/512/1024/2048/4096 tokens | | ------------------------------------------------ | ---------------- | ----------------- | | OmniSVG-3B | 17G | 4.08/8.68/18.07/37.51/82.70 seconds | Note: The inference time shown here is measured per OmniSVG SVG tokens, while the inference time reported in our paper is measured per XML code tokens for fair comparison with baseline methods. Download the model weights from Hugging Face and place them in the `./pretrainedmodels/OmniSVG-3B` directory. Execute the following command to run inference on your images: We provide an interactive generation interface using Gradio: 5. License OmniSVG is licensed under the Apache License 2.0, while MMSVG dataset is under Creative Commons Attribution Non Commercial Share Alike 4.0 License. You can find the license files in the respective github and HuggingFace repositories. Acknowledgments We thank the following excellent open-source works: IconShop: is the first advanced work that leverages LLMs to generate monochrome, icon-level SVGs. We referred to its parametric implementation. Here is the list of highly related concurrent works: LLM4SVG: treats SVG coordinates as number strings and predicts decimal part for higher spatial accuracy. StarVector: equips LLM with an image encoder for Image-to-SVG generation.