Starred repositories
A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training
Efficient Actively Secure DPF and RAM-based 2PC with One-Bit Leakage
Training library for Megatron-based models with bi-directional Hugging Face conversion capability
Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support
TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs
Official repository for the paper "MICo-150K: A Comprehensive Dataset for Multi-Image Composition".
CaptionQA: Is Your Caption as Useful as the Image Itself?
PyTorch building blocks for the OLMo ecosystem
Paper list for Efficient Reasoning.
TraceRL & TraDo-8B: Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models
A lightweight Inference Engine built for block diffusion models
Structuring Hour-Long Videos into Navigable Chapters and Hierarchical Summaries
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
SDAR (Synergy of Diffusion and AutoRegression), a large diffusion language model（1.7B, 4B, 8B, 30B）
This is the official repository for the ICCV 2025 paper ReCoT: Reflective Self-Correction Training for Mitigating Confirmation Bias in Large Vision-Language Models
Sequential Diffusion Language Model (SDLM) enhances pre-trained autoregressive language models by adaptively determining generation length and maintaining KV-cache compatibility, achieving high eff…
Minimalistic large language model 3D-parallelism training
We have summarised all 3D anomaly detection methods and datasets (still updating). 多模态，点云和姿势无关异常检测的综述仓库（持续更新）
48
Updated Jan 14, 2026
VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo
The most open diffusion language model for code generation — releasing pretraining, evaluation, inference, and checkpoints.
The official github repo for "Diffusion Language Models are Super Data Learners".
GPU-optimized framework for training diffusion language models at any scale. The backend of Quokka, Super Data Learners, and OpenMoE 2 training.
Ongoing research training transformer models at scale
PyTorch native quantization and sparsity for training and inference
Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"