MiniMax M2.1, a SOTA model for real-world dev & agents.

Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction

Towards Scalable Pre-training of Visual Tokenizers for Generation

A high-throughput and memory-efficient inference and serving engine for LLMs

Official PyTorch Code for "OmniAID: Decoupling Semantic and Artifacts for Universal AI-Generated Image Detection in the Wild".

Semantics Lead the Way: Harmonizing Semantic and Texture Modeling with Asynchronous Latent Diffusion

[ECCV2024] Video Foundation Models & Data for Multimodal Understanding

Paper Debugger is the best overleaf companion

A minimal PyTorch re-implementation of Qwen3 VL with a fancy CLI

HunyuanVideo-1.5: A leading lightweight video generation model

PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

Official Implementation of "MMaDA-Parallel: Multimodal Large Diffusion Language Models for Thinking-Aware Editing and Generation"

The NVIDIA® Tools Extension SDK (NVTX) is a C-based Application Programming Interface (API) for annotating events, code ranges, and resources in your applications.

State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

Triton implementation of FlashAttention2 that adds Custom Masks.

[ICCV2023 Oral] Unmasked Teacher: Towards Training-Efficient Video Foundation Models

Official implementation of paper "VMoBA: Mixture-of-Block Attention for Video Diffusion Models"

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

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

A professional cross-platform SSH/Sftp/Shell/Telnet/Tmux/Serial terminal.

Cyberduck is a libre FTP, SFTP, WebDAV, Amazon S3, Backblaze B2, Microsoft Azure & OneDrive and OpenStack Swift file transfer client for Mac and Windows.

MiniMax-M2, a model built for Max coding & agentic workflows.

On the Hidden Mystery of OCR in Large Multimodal Models (OCRBench)

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

NipaPlay-Reload 是一个现代化的跨平台本地视频播放器，支持 Windows、macOS、Linux、Android 和 iOS。集成了弹幕显示、多格式字幕支持、多音频轨道切换，新番查看等功能，支持挂载Emby/Jellyfin媒体库。采用 Flutter 开发，提供统一的用户体验。

Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"