LottieGPT: Tokenizing Vector Animation for Autoregressive Generation

2026-04-13

AI Summary

Existing video generation models struggle to produce resolution-independent, semantically structured, and motion-editable vector animations. This work proposes the first end-to-end autoregressive generative framework tailored for vector animation, leveraging the Lottie standard to design an efficient tokenizer that encodes geometric primitives, transformations, and keyframes into compact, semantically aligned token sequences. Integrated with the Qwen-VL multimodal large language model, the framework enables text- or image-to-vector-animation generation. Key contributions include constructing the largest-scale Lottie animation dataset to date, achieving significant token-length compression while preserving structural fidelity, demonstrating strong generalization across diverse artistic styles, and outperforming state-of-the-art methods on SVG generation tasks.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Despite rapid progress in video generation, existing models are incapable of producing vector animation, a dominant and highly expressive form of multimedia on the Internet. Vector animations offer resolution-independence, compactness, semantic structure, and editable parametric motion representations, yet current generative models operate exclusively in raster space and thus cannot synthesize them. Meanwhile, recent advances in large multimodal models demonstrate strong capabilities in generating structured data such as slides, 3D meshes, LEGO sequences, and indoor layouts, suggesting that native vector animation generation may be achievable. In this work, we present the first framework for tokenizing and autoregressively generating vector animations. We adopt Lottie, a widely deployed JSON-based animation standard, and design a tailored Lottie Tokenizer that encodes layered geometric primitives, transforms, and keyframe-based motion into a compact and semantically aligned token sequence. To support large-scale training, we also construct LottieAnimation-660K, the largest and most diverse vector animation dataset to date, consisting of 660k real-world Lottie animation and 15M static Lottie image files curated from broad Internet sources. Building upon these components, we finetune Qwen-VL to create LottieGPT, a native multimodal model capable of generating coherent, editable vector animations directly from natural language or visual prompts. Experiments show that our tokenizer dramatically reduces sequence length while preserving structural fidelity, enabling effective autoregressive learning of dynamic vector content. LottieGPT exhibits strong generalization across diverse animation styles and outperforms previous state-of-the-art models on SVG generation (a special case of single-frame vector animation).

Problem

• vector animation
• autoregressive generation
• Lottie
• raster space
• structured data

Innovation

• vector animation
• Lottie
• autoregressive generation
• structured tokenization
• multimodal generative model