NVIDIA / TransformerEngineLinks
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.
☆2,883Updated this week
Alternatives and similar repositories for TransformerEngine
Users that are interested in TransformerEngine are comparing it to the libraries listed below
Sorting:
- FlashInfer: Kernel Library for LLM Serving☆3,982Updated last week
- PyTorch native quantization and sparsity for training and inference☆2,489Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,536Updated last year
- Transformer related optimization, including BERT, GPT☆6,338Updated last year
- A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. …☆1,486Updated last week
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,324Updated 3 months ago
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,185Updated 2 months ago
- Tile primitives for speedy kernels☆2,859Updated last week
- Pipeline Parallelism for PyTorch☆780Updated last year
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,072Updated 4 months ago
- PyTorch extensions for high performance and large scale training.☆3,384Updated 6 months ago
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,841Updated this week
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆2,180Updated this week
- Minimalistic large language model 3D-parallelism training☆2,294Updated 2 months ago
- SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX R…☆2,521Updated this week
- NCCL Tests☆1,313Updated last week
- Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".☆2,212Updated last year
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆1,931Updated this week
- Puzzles for learning Triton☆2,090Updated 11 months ago
- Distributed Compiler based on Triton for Parallel Systems☆1,206Updated 2 weeks ago
- The Triton TensorRT-LLM Backend☆904Updated this week
- A pytorch quantization backend for optimum☆1,004Updated 2 weeks ago
- 🚀 Efficient implementations of state-of-the-art linear attention models☆3,661Updated this week
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,261Updated 5 months ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆3,837Updated this week
- A PyTorch native platform for training generative AI models☆4,645Updated this week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,161Updated 2 months ago
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆883Updated last week
- A PyTorch Native LLM Training Framework☆883Updated last month
- Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 using FP8/NVFP4/MXFP4☆934Updated last month