auto-regressive-model 开源项目推荐（52 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！

buaacyw /MeshAnything

[ICLR 2025] From anything to mesh like human artists. Official impl. of "MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers"

FoundationVision /LlamaGen

自回归模型击败扩散：为可扩展图像生成寻找Llama

FoundationVision /Infinity

[CVPR 2025 Oral] Infinity ∞：扩展按位自回归建模以实现高分辨率图像合成

omerbsezer /Generative_Models_Tutorial_with_Demo

生成模型教程与演示：贝叶斯分类器采样、变分自编码器（VAE）、生成对抗网络（GAN）、流行 GAN 架构、自回归模型、重要生成模型论文、课程等。

FoundationVision /OmniTokenizer

[NeurIPS 2024]OmniTokenizer: one model and one weight for image-video joint tokenization.

SunzeY /X-Prompt

X-Prompt 的官方实现：面向自回归视觉语言基础模型的通用上下文图像生成

SOTAMak1r /Infinite-Forcing

Infinite-Forcing: Towards Infinite-Long Video Generation

hp-l33 /AiM

Official PyTorch Implementation of "Scalable Autoregressive Image Generation with Mamba"

pactools /pactools

相位-振幅耦合（PAC）工具箱

czg1225 /CoDe

[CVPR 2025] CoDe: Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient

MKJia /MGVQ

[Arxiv'25] MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group Quantization

StargazerX0 /ScaleKV

[NeurIPS 2025] ScaleKV: Memory-Efficient Visual Autoregressive Modeling with Scale-Aware KV Cache Compression

zhanghm1995 /Awesome-VAR

A curated list of resources focused on Visual AutoRegressive Modeling, makes GPT-style AR models surpass diffusion transformers in image generation.

miyosuda /heartrate-monitor

Heart rate variability (HRV) analysis tool to detect autonomic nerve state

cam1681 /CS25_Large_Foundation_model

Lecture slides for the Course "Large Foundation Models: Mathematics, Algorithms, and Applications"

aimagelab /Emuru-autoregressive-text-img

Official PyTorch implementation for "Zero-Shot Styled Text Image Generation, but Make It Autoregressive" (CVPR25)

AlgonetLabs /Cable

Context-aware Biases for Length Extrapolation

LlamaGenAI /llamagen

AI Comic Factory - Generate Comics with AI, 🦙 Llama for Scalable Anime Generation, Image Generation, Comic Generation and Game Generation

hogru /MolReactGen

Auto-regressive causal language model for molecule (SMILES) and reaction template (SMARTS) generation based on the Hugging Face implementation of OpenAI's GPT-2 transformer decoder model