NVIDIA-NeMo / AutomodelLinks
DTensor-native pretraining and fine-tuning for LLMs/VLMs with day-0 Hugging Face support, GPU-acceleration, and memory efficiency.
β84Updated this week
Alternatives and similar repositories for Automodel
Users that are interested in Automodel are comparing it to the libraries listed below
Sorting:
- π Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flashβ¦β268Updated 2 months ago
- Load compute kernels from the Hubβ290Updated last week
- π Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.β213Updated this week
- Training library for Megatron-based modelsβ109Updated this week
- β122Updated last year
- Flash-Muon: An Efficient Implementation of Muon Optimizerβ189Updated 3 months ago
- β126Updated 4 months ago
- Megatron's multi-modal data loaderβ250Updated last week
- Memory optimized Mixture of Expertsβ67Updated 2 months ago
- Triton-based implementation of Sparse Mixture of Experts.β242Updated this week
- ring-attention experimentsβ152Updated 11 months ago
- ArcticTraining is a framework designed to simplify and accelerate the post-training process for large language models (LLMs)β219Updated this week
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Trainingβ216Updated last year
- Odysseus: Playground of LLM Sequence Parallelismβ77Updated last year
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS β¦β60Updated 11 months ago
- This repository contains the experimental PyTorch native float8 training UXβ224Updated last year
- PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"β73Updated 3 weeks ago
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mindβ¦β161Updated 2 weeks ago
- Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.β98Updated 2 weeks ago
- The evaluation framework for training-free sparse attention in LLMsβ98Updated 3 months ago
- β112Updated last year
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top ofβ¦β146Updated last year
- Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)β414Updated this week
- FlexAttention based, minimal vllm-style inference engine for fast Gemma 2 inference.β280Updated last month
- A scalable asynchronous reinforcement learning implementation with in-flight weight updates.β151Updated this week
- A tool to configure, launch and manage your machine learning experiments.β195Updated this week
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.β89Updated 2 months ago
- β221Updated 7 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clustersβ130Updated 10 months ago
- DPO, but faster πβ44Updated 10 months ago