lhb8125 / Megatron-LMLinks
Ongoing research training transformer models at scale
☆18Updated last week
Alternatives and similar repositories for Megatron-LM
Users that are interested in Megatron-LM are comparing it to the libraries listed below
Sorting:
- [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…☆65Updated last year
- ☆100Updated last year
- Pipeline Parallelism Emulation and Visualization☆68Updated 4 months ago
- A lightweight design for computation-communication overlap.☆181Updated last week
- nnScaler: Compiling DNN models for Parallel Training☆118Updated last month
- ☆107Updated 5 months ago
- Sequence-level 1F1B schedule for LLMs.☆32Updated last month
- PyTorch bindings for CUTLASS grouped GEMM.☆154Updated last week
- ☆148Updated 7 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆76Updated last year
- ☆141Updated 9 months ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆138Updated last month
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆112Updated 5 months ago
- ☆150Updated 9 months ago
- A baseline repository of Auto-Parallelism in Training Neural Networks☆147Updated 3 years ago
- ☆137Updated 10 months ago
- High performance Transformer implementation in C++.☆138Updated 9 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated last month
- PyTorch bindings for CUTLASS grouped GEMM.☆124Updated 4 months ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆67Updated 7 months ago
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆80Updated 11 months ago
- ☆56Updated 3 months ago
- Examples of CUDA implementations by Cutlass CuTe☆241Updated 3 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆282Updated last year
- Allow torch tensor memory to be released and resumed later☆150Updated this week
- ☆139Updated last year
- High Performance Grouped GEMM in PyTorch☆31Updated 3 years ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆221Updated 2 years ago
- ☆307Updated 3 weeks ago
- PyTorch distributed training acceleration framework☆53Updated 2 months ago