OpenSQZ / MegatronAppLinks
Toolchain built around the Megatron-LM for Distributed Training
☆61Updated 3 weeks ago
Alternatives and similar repositories for MegatronApp
Users that are interested in MegatronApp are comparing it to the libraries listed below
Sorting:
- Allow torch tensor memory to be released and resumed later☆115Updated 3 weeks ago
- A simple calculation for LLM MFU.☆44Updated 5 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆138Updated this week
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆103Updated this week
- ByteCheckpoint: An Unified Checkpointing Library for LFMs☆238Updated last month
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆214Updated last year
- ☆93Updated 5 months ago
- Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.☆64Updated last week
- Utility scripts for PyTorch (e.g. Memory profiler that understands more low-level allocations such as NCCL)☆50Updated 3 weeks ago
- Odysseus: Playground of LLM Sequence Parallelism☆76Updated last year
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆201Updated this week
- ☆78Updated 4 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆110Updated 3 months ago
- Pipeline Parallelism Emulation and Visualization☆63Updated 2 months ago
- ☆284Updated last week
- A lightweight reinforcement learning framework that integrates seamlessly into your codebase, empowering developers to focus on algorithm…☆61Updated last week
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆344Updated this week
- Training library for Megatron-based models☆54Updated this week
- ☆121Updated last year
- A collection of memory efficient attention operators implemented in the Triton language.☆278Updated last year
- ☆123Updated 3 months ago
- ☆42Updated last year
- Implementation for FP8/INT8 Rollout for RL training without performence drop.☆184Updated this week
- nnScaler: Compiling DNN models for Parallel Training☆118Updated this week
- ☆146Updated 5 months ago
- Zero Bubble Pipeline Parallelism☆423Updated 3 months ago
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆234Updated 3 weeks ago
- ☆196Updated 3 months ago
- [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…☆61Updated last year
- A minimal implementation of vllm.☆52Updated last year