Expert Kit is an efficient foundation of Expert Parallelism (EP) for MoE model Inference on heterogenous hardware

An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

利用AI大模型，一键生成高清短视频 Generate short videos with one click using AI LLM.

Accelerating the development of large multimodal models (LMMs) with one-click evaluation module - lmms-eval.

Megatron's multi-modal data loader

Puzzles for learning Triton

MoBA: Mixture of Block Attention for Long-Context LLMs

🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"

New repo collection for NVIDIA Cosmos: https://github.com/nvidia-cosmos

北航研究生学位论文模板（Word+LaTeX）.

[ECCV 2024] FreeInit: Bridging Initialization Gap in Video Diffusion Models

「硬地骇客 - 两个月 $12000 ARR 实践之路」是由 硬地骇客 团队编著，本书是关于 Podwise 产品历程的忠实记录：内容包含 灵感 - 构建 - 发布 - 增长 - 复盘 五个章节。如果你觉得一个人读不够过瘾，欢迎加入「硬地骇客」官方知识星球与专家们一起讨论！Podwise 的故事才刚刚开始，我们也将在星球持续分享我们的认知，成功可能无法复制，但失败一定可以借鉴。现在就点击下方…

Nodes for image juxtaposition for Flux in ComfyUI

Rectified Flow Inversion (RF-Inversion) - ICLR 2025

Platform for Situated Intelligence

An open-source RAG-based tool for chatting with your documents.

Open Source alternative to Algolia + Pinecone and an Easier-to-Use alternative to ElasticSearch ⚡ 🔍 ✨ Fast, typo tolerant, in-memory fuzzy Search Engine for building delightful search experiences

Official inference repo for FLUX.1 models

The calflops is designed to calculate FLOPs、MACs and Parameters in all various neural networks, such as Linear、 CNN、 RNN、 GCN、Transformer(Bert、LlaMA etc Large Language Model)

Memray is a memory profiler for Python

A feature-rich command-line audio/video downloader

SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

[CVPR 2025] StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text

An Open-source Toolkit for LLM Development

Implementation of Lumiere, SOTA text-to-video generation from Google Deepmind, in Pytorch

Official Code for Stable Cascade

Open-Sora: Democratizing Efficient Video Production for All

Official implementation of "DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents"

[TMLR 2025] Latte: Latent Diffusion Transformer for Video Generation.