intelligent-machine-learning / atorchLinks
An industrial extension library of pytorch to accelerate large scale model training
☆49Updated 2 months ago
Alternatives and similar repositories for atorch
Users that are interested in atorch are comparing it to the libraries listed below
Sorting:
- [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training☆244Updated 3 months ago
- ☆149Updated 8 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆283Updated last year
- ☆97Updated 7 months ago
- An easy-to-use package for implementing SmoothQuant for LLMs☆107Updated 7 months ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆145Updated 2 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆164Updated 3 weeks ago
- ☆64Updated 2 weeks ago
- A lightweight reinforcement learning framework that integrates seamlessly into your codebase, empowering developers to focus on algorithm…☆68Updated 2 months ago
- Distributed IO-aware Attention algorithm☆21Updated last month
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆460Updated this week
- Odysseus: Playground of LLM Sequence Parallelism☆78Updated last year
- Official implementation of ICML 2024 paper "ExCP: Extreme LLM Checkpoint Compression via Weight-Momentum Joint Shrinking".☆48Updated last year
- FlagScale is a large model toolkit based on open-sourced projects.☆398Updated last week
- Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training☆217Updated last year
- ByteCheckpoint: An Unified Checkpointing Library for LFMs☆251Updated 4 months ago
- ☆429Updated 2 months ago
- A quantization algorithm for LLM☆145Updated last year
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆586Updated 3 weeks ago
- ☆129Updated 10 months ago
- ☆79Updated 6 months ago
- InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencie…☆411Updated 2 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆126Updated 5 months ago
- Built upon Megatron-Deepspeed and HuggingFace Trainer, EasyLLM has reorganized the code logic with a focus on usability. While enhancing …☆48Updated last year
- ☆79Updated last year
- Pipeline Parallelism Emulation and Visualization☆70Updated 4 months ago
- [ICLR 2025] PEARL: Parallel Speculative Decoding with Adaptive Draft Length☆125Updated last week
- ☆120Updated 2 months ago
- Zero Bubble Pipeline Parallelism☆433Updated 6 months ago
- 青稞Talk☆157Updated 2 weeks ago