NVIDIA / Megatron-EnergonLinks
Megatron's multi-modal data loader
β213Updated 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¦β253Updated last week
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Trainingβ210Updated last week
- π₯ A minimal training framework for scaling FLA modelsβ178Updated 2 weeks ago
- This repository contains the experimental PyTorch native float8 training UXβ224Updated 10 months ago
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Trainingβ210Updated 10 months ago
- Scalable toolkit for efficient model reinforcementβ448Updated this week
- ByteCheckpoint: An Unified Checkpointing Library for LFMsβ219Updated 2 months ago
- π Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.β204Updated this week
- Ring attention implementation with flash attentionβ789Updated 2 weeks ago
- ring-attention experimentsβ144Updated 8 months ago
- Implementation of π Ring Attention, from Liu et al. at Berkeley AI, in Pytorchβ521Updated last month
- Efficient LLM Inference over Long Sequencesβ378Updated 3 weeks ago
- Zero Bubble Pipeline Parallelismβ399Updated last month
- Large Context Attentionβ716Updated 5 months ago
- Applied AI experiments and examples for PyTorchβ277Updated 3 weeks ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inferenceβ520Updated last month
- β193Updated 4 months ago
- Triton implementation of FlashAttention2 that adds Custom Masks.β120Updated 10 months ago
- [ICML 2024] CLLMs: Consistency Large Language Modelsβ395Updated 7 months ago
- VeOmni: Scaling any Modality Model Training to any Accelerators with PyTorch native Training Frameworkβ355Updated last month
- Flash-Muon: An Efficient Implementation of Muon Optimizerβ131Updated last week
- PyTorch bindings for CUTLASS grouped GEMM.β100Updated 3 weeks ago
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Headsβ466Updated 4 months ago
- PyTorch bindings for CUTLASS grouped GEMM.β127Updated 5 months ago
- Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"β233Updated 2 weeks ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.β167Updated this week
- Triton-based implementation of Sparse Mixture of Experts.β219Updated 7 months ago
- β114Updated 3 weeks ago
- Efficient triton implementation of Native Sparse Attention.β168Updated last month
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantizationβ359Updated 10 months ago