NVIDIA / TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.
☆2,086Updated this week
Alternatives and similar repositories for TransformerEngine:
Users that are interested in TransformerEngine are comparing it to the libraries listed below
- FlashInfer: Kernel Library for LLM Serving☆1,797Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,316Updated 6 months ago
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆2,669Updated last week
- Pipeline Parallelism for PyTorch☆736Updated 4 months ago
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,957Updated 3 weeks ago
- PyTorch extensions for high performance and large scale training.☆3,232Updated this week
- TensorRT Model Optimizer is a unified library of state-of-the-art model optimization techniques such as quantization, pruning, distillati…☆667Updated last week
- Transformer related optimization, including BERT, GPT☆5,981Updated 9 months ago
- Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackab…☆1,549Updated 11 months ago
- PyTorch native quantization and sparsity for training and inference☆1,753Updated this week
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,749Updated this week
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆1,942Updated last month
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,179Updated 3 months ago
- Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".☆1,998Updated 9 months ago
- A Python-level JIT compiler designed to make unmodified PyTorch programs faster.☆1,016Updated 9 months ago
- Tile primitives for speedy kernels☆1,923Updated this week
- SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX R…☆2,298Updated this week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,354Updated 9 months ago
- Minimalistic large language model 3D-parallelism training☆1,386Updated this week
- The Triton TensorRT-LLM Backend☆745Updated last week
- A pytorch quantization backend for optimum☆865Updated last week
- Microsoft Automatic Mixed Precision Library☆549Updated 3 months ago
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆1,885Updated 2 weeks ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆680Updated 4 months ago
- 🚀 Efficient implementations of state-of-the-art linear attention models in Pytorch and Triton☆1,669Updated this week
- Tutel MoE: An Optimized Mixture-of-Experts Implementation☆746Updated this week
- GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection☆1,481Updated 2 months ago
- A throughput-oriented high-performance serving framework for LLMs☆692Updated 3 months ago
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆714Updated this week