NVIDIA / Megatron-EnergonLinks
Megatron's multi-modal data loader
β227Updated this week
Alternatives and similar repositories for Megatron-Energon
Users that are interested in Megatron-Energon 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¦β256Updated 2 weeks ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Trainingβ217Updated last month
- π₯ A minimal training framework for scaling FLA modelsβ201Updated last month
- Efficient triton implementation of Native Sparse Attention.β177Updated 2 months ago
- VeOmni: Scaling any Modality Model Training to any Accelerators with PyTorch native Training Frameworkβ379Updated this week
- Scalable toolkit for efficient model reinforcementβ532Updated this week
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Trainingβ212Updated 11 months ago
- Triton implementation of FlashAttention2 that adds Custom Masks.β126Updated 11 months ago
- Efficient LLM Inference over Long Sequencesβ385Updated last month
- ByteCheckpoint: An Unified Checkpointing Library for LFMsβ226Updated 2 weeks ago
- Flash-Muon: An Efficient Implementation of Muon Optimizerβ144Updated last month
- Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"β305Updated 2 weeks ago
- Implementation of π Ring Attention, from Liu et al. at Berkeley AI, in Pytorchβ532Updated 2 months ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Headsβ477Updated 5 months ago
- Large Context Attentionβ720Updated 6 months ago
- π Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.β207Updated this week
- Microsoft Automatic Mixed Precision Libraryβ616Updated 9 months ago
- Ring attention implementation with flash attentionβ818Updated this week
- This repository contains the experimental PyTorch native float8 training UXβ224Updated 11 months ago
- [ICML 2024] CLLMs: Consistency Large Language Modelsβ397Updated 8 months ago
- Triton-based implementation of Sparse Mixture of Experts.β225Updated 7 months ago
- ring-attention experimentsβ145Updated 9 months ago
- β122Updated last month
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inferenceβ532Updated last week
- Low-bit optimizers for PyTorchβ130Updated last year
- [CoLM'25] The official implementation of the paper <MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression>β140Updated 2 weeks ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMsβ181Updated last month
- Odysseus: Playground of LLM Sequence Parallelismβ72Updated last year
- [ICLR2025] Codebase for "ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing", built on Megatron-LM.β83Updated 7 months ago
- [ICML 2025] XAttention: Block Sparse Attention with Antidiagonal Scoringβ206Updated 2 weeks ago