kwai / Megatron-Kwai
[USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Parallelism
☆51Updated 7 months ago
Alternatives and similar repositories for Megatron-Kwai:
Users that are interested in Megatron-Kwai are comparing it to the libraries listed below
- PyTorch bindings for CUTLASS grouped GEMM.☆103Updated 2 months ago
- ☆85Updated 4 months ago
- ☆120Updated last week
- ☆84Updated 6 months ago
- ☆72Updated 3 years ago
- A baseline repository of Auto-Parallelism in Training Neural Networks☆143Updated 2 years ago
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆93Updated last year
- nnScaler: Compiling DNN models for Parallel Training☆101Updated 3 weeks ago
- PyTorch distributed training acceleration framework☆44Updated last month
- ☆139Updated 10 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆250Updated 9 months ago
- Sequence-level 1F1B schedule for LLMs.☆16Updated 2 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆35Updated 2 weeks ago
- ☆79Updated 3 months ago
- ☆143Updated 2 months ago
- A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of …☆144Updated 8 months ago
- Since the emergence of chatGPT in 2022, the acceleration of Large Language Model has become increasingly important. Here is a list of pap…☆230Updated last week
- Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines.☆60Updated last year
- ☆52Updated 11 months ago
- ATC23 AE☆45Updated last year
- Dynamic Memory Management for Serving LLMs without PagedAttention☆308Updated 3 weeks ago
- ☆127Updated 2 months ago
- Curated collection of papers in MoE model inference☆93Updated 3 weeks ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆107Updated 3 weeks ago
- This repository is established to store personal notes and annotated papers during daily research.☆112Updated last week
- High performance Transformer implementation in C++.☆105Updated last month
- Zero Bubble Pipeline Parallelism☆361Updated last week
- flash attention tutorial written in python, triton, cuda, cutlass☆296Updated 2 months ago
- ☆29Updated 7 months ago
- ☆63Updated 3 months ago