Pinned Repositories
Megatron-LM
Ongoing research training transformer models at scale
TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.
Adaptation-SCR
Official ICCV code of Text-Driven Generative Domain Adaptation with Spectral Consistency Regularization
awesome-image-synthesis
Collection of papers in image synthesis with paper graph.
awesome-poe-bots
Useful chatbots of poe.
Image-Generation-models
Generative models (GAN, VAE, Diffusion Models, Autoregressive Models) implemented with Pytorch, Pytorch_lightning and hydra.
Image-Translation-Models
PP-Schedule-Visualization
Pipeline Parallelism Emulation and Visualization
PyTorch-Memory-Profiler
stable-dreambooth
Dreambooth implementation based on Stable Diffusion with minimal code.
Victarry's Repositories
Victarry/stable-dreambooth
Dreambooth implementation based on Stable Diffusion with minimal code.
Victarry/PP-Schedule-Visualization
Pipeline Parallelism Emulation and Visualization
Victarry/Image-Generation-models
Generative models (GAN, VAE, Diffusion Models, Autoregressive Models) implemented with Pytorch, Pytorch_lightning and hydra.
Victarry/PyTorch-Memory-Profiler
Victarry/awesome-image-synthesis
Collection of papers in image synthesis with paper graph.
Victarry/awesome-poe-bots
Useful chatbots of poe.
Victarry/Image-Translation-Models
Victarry/Adaptation-SCR
Official ICCV code of Text-Driven Generative Domain Adaptation with Spectral Consistency Regularization
Victarry/Awesome-AI-Scaling
Victarry/PMC-Video-Animation
Victarry/Megatron-MoE-ModelZoo
Best practices for testing advanced Mixtral, DeepSeek, and Qwen series MoE models using Megatron Core MoE.
Victarry/Victarry.github.io
Victarry/xDiT
xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) on multi-GPU Clusters
Victarry/Adaptation-SCR-website
Victarry/Cosmos
Cosmos is a world model development platform that consists of world foundation models, tokenizers and video processing pipeline to accelerate the development of Physical AI at Robotics & AV labs. Cosmos is purpose built for physical AI. The Cosmos repository will enable end users to run the Cosmos models, run inference scripts and generate videos.
Victarry/CSAPP-labs
Victarry/Dash-User-Contributions
Dash repo of user contributed docsets
Victarry/GPU-Puzzles
Solve puzzles. Learn CUDA.
Victarry/happy-new-year
Victarry/how-to-optim-algorithm-in-cuda
how to optimize some algorithm in cuda.
Victarry/Huarongdao
华容道游戏, 采用libgdx框架
Victarry/Megatron-LM
Ongoing research training transformer models at scale
Victarry/ml-engineering
Machine Learning Engineering Open Book
Victarry/NeMo
A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)
Victarry/ScholArxiv
ScholArxiv is an open-source aesthetic and minimal app that allows users to search, read, bookmark, share, download and view summaries of academic papers from the arXiv repository.
Victarry/torchtitan
A native PyTorch Library for large model training
Victarry/TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.
Victarry/tutel
Tutel MoE: An Optimized Mixture-of-Experts Implementation
Victarry/unsloth
Finetune Llama 3.1, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
Victarry/Victarry