Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 using FP8/NVFP4/MXFP4
☆1,008Jul 21, 2026Updated 2 weeks ago
Alternatives and similar repositories for Tutel
Users that are interested in Tutel are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- A fast MoE impl for PyTorch☆1,858Feb 10, 2025Updated last year
- ☆1,585Mar 25, 2026Updated 4 months ago
- ☆726Updated this week
- Microsoft Collective Communication Library☆394Sep 20, 2023Updated 2 years ago
- PyTorch Re-Implementation of "The Sparsely-Gated Mixture-of-Experts Layer" by Noam Shazeer et al. https://arxiv.org/abs/1701.06538☆1,247Apr 19, 2024Updated 2 years ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- A collection of AWESOME things about mixture-of-experts☆1,285Dec 8, 2024Updated last year
- PyTorch extensions for high performance and large scale training.☆3,409Apr 26, 2025Updated last year
- ☆92Apr 2, 2022Updated 4 years ago
- A Pytorch implementation of Sparsely-Gated Mixture of Experts, for massively increasing the parameter count of language models☆868Sep 13, 2023Updated 2 years ago
- Training and serving large-scale neural networks with auto parallelization.☆3,181Dec 9, 2023Updated 2 years ago
- Compiler for Dynamic Neural Networks☆45Nov 13, 2023Updated 2 years ago
- A family of open-sourced Mixture-of-Experts (MoE) Large Language Models☆1,694Mar 8, 2024Updated 2 years ago
- Synthesizer for optimal collective communication algorithms☆126Apr 8, 2024Updated 2 years ago
- ☆113Aug 26, 2024Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,898Updated this week
- A curated reading list of research in Mixture-of-Experts(MoE).☆669Oct 30, 2024Updated last year
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,484Updated this week
- Transformer related optimization, including BERT, GPT☆6,445Mar 27, 2024Updated 2 years ago
- Ongoing research training transformer models at scale☆17,378Updated this week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,258Aug 14, 2025Updated 11 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆281Oct 3, 2025Updated 10 months ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,352Aug 28, 2025Updated 11 months ago
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆1,036Mar 3, 2026Updated 5 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- This package implements THOR: Transformer with Stochastic Experts.☆64Oct 7, 2021Updated 4 years ago
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆546Updated this week
- Microsoft Automatic Mixed Precision Library☆635Dec 1, 2025Updated 8 months ago
- Ring attention implementation with flash attention☆1,045Sep 10, 2025Updated 10 months ago
- An experimental parallel training platform☆57Mar 25, 2024Updated 2 years ago
- ⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)☆1,004Dec 6, 2024Updated last year
- Distributed Compiler based on Triton for Parallel Systems☆1,512Updated this week
- nnScaler: Compiling DNN models for Parallel Training☆135Jul 2, 2026Updated last month
- PyTorch bindings for CUTLASS grouped GEMM.☆192Apr 8, 2026Updated 4 months ago
- End-to-end encrypted email - Proton Mail • AdSpecial offer: 40% Off Yearly / 80% Off First Month. All Proton services are open source and independently audited for security.
- A high-performance distributed deep learning system targeting large-scale and automated distributed training.☆340Dec 13, 2025Updated 7 months ago
- Development repository for the Triton language and compiler☆19,908Updated this week
- NCCL Profiling Kit☆156Jul 1, 2024Updated 2 years ago
- Fast and memory-efficient exact attention☆24,658Updated this week
- ☆143Jan 30, 2025Updated last year
- A throughput-oriented high-performance serving framework for LLMs☆974Mar 29, 2026Updated 4 months ago
- Pipeline Parallelism for PyTorch☆786Aug 21, 2024Updated last year