autoregressive-models 开源项目推荐（187 个）

FoundationVision /VAR

[NeurIPS 2024年最佳论文奖][GPT击败扩散] [视觉生成]官方实施。🔥“视觉自回归建模：通过下一规模预测的可扩展图像生成”。一个 * 超简单、用户友好且最先进的 * 代码库，用于自回归图像生成！

FoundationVision /Infinity

[CVPR 2025 Oral]Infinity ∞ : Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

pdebench /PDEBench

PDEBench: An Extensive Benchmark for Scientific Machine Learning

thu-ml /Causal-Forcing

[ICML 2026] Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation" & Causal Forcing++

FoundationVision /InfinityStar

[NeurIPS 2025 Oral]Infinity⭐️: Unified Spacetime AutoRegressive Modeling for Visual Generation

baaivision /NOVA

[ICLR 2025] Autoregressive Video Generation without Vector Quantization

FoundationVision /Liquid

(Accepted by IJCV) Liquid: Language Models are Scalable and Unified Multi-modal Generators

FoundationVision /UniTok

[NeurIPS 2025 Spotlight] A Unified Tokenizer for Visual Generation and Understanding

zhao-kun /VibeVoiceFusion

VibeVoiceFusion 是一个全栈、多扬声器语音生成网络系统，具备 LoRA 微调、批量生成和显存优化功能。基于Microsoft的VibeVoice（AR + 扩散架构）

nv-tlabs /ATISS

Code for "ATISS: Autoregressive Transformers for Indoor Scene Synthesis", NeurIPS 2021

lxa9867 /ImageFolder

High-performance Image Tokenizers for VAR and AR

Yi-Shi94 /AMDM

Interactive Character Control with Auto-Regressive Motion Diffusion Models

WangHelin1997 /SSR-Speech

SSR-Speech: Towards Stable, Safe and Robust Zero-shot Speech Editing and Synthesis

showlab /D-AR

the official repo for "D-AR: Diffusion via Autoregressive Models"

XianfengWu01 /LightGen

An Efficient Text-to-Image Generation Pretrain Pipeline

fajieyuan /WSDM2019-nextitnet

Generative model for sequential recommendation based on Convolution Neural Networks (CNN))

MILab-PKU /MVAR

Offical implementation of "Auto-Regressively Generating Multi-View Consistent Images". (ICCV 2025)

KumapowerLIU /AvatarPointillist

[CVPR 2026]Official PyTorch implementation of AvatarPointillist: AutoRegressive 4D Gaussian Avatarization.

Cornell-VAILab /Raster2Seq

Raster2Seq: Polygon Sequence Generation for Floorplan Reconstruction (SIGGRAPH 2026)

FoundationVision /BitVAE

official training and inference code of bitwise tokenizer