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.
☆2,627Updated last 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☆3,524Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,471Updated last year
- PyTorch native quantization and sparsity for training and inference☆2,245Updated last week
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,195Updated 3 weeks ago
- Transformer related optimization, including BERT, GPT☆6,270Updated last year
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,815Updated last week
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,134Updated 3 weeks ago
- A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. …☆1,103Updated this week
- Tile primitives for speedy kernels☆2,570Updated last week
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,045Updated last month
- Pipeline Parallelism for PyTorch☆774Updated 11 months ago
- PyTorch extensions for high performance and large scale training.☆3,355Updated 3 months ago
- Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".☆2,161Updated last year
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆1,506Updated last week
- 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…☆1,577Updated last year
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,270Updated 5 months ago
- NCCL Tests☆1,214Updated 3 weeks ago
- Minimalistic large language model 3D-parallelism training☆2,121Updated last month
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,406Updated last year
- SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX R…☆2,472Updated this week
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3.☆1,462Updated last week
- The Triton TensorRT-LLM Backend☆878Updated this week
- A Python-level JIT compiler designed to make unmodified PyTorch programs faster.☆1,056Updated last year
- Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 FP8/NVFP4/MXFP4☆885Updated last week
- A pytorch quantization backend for optimum☆981Updated last month
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆1,663Updated last week
- Microsoft Automatic Mixed Precision Library☆616Updated 10 months ago
- Flash Attention in ~100 lines of CUDA (forward pass only)☆903Updated 7 months ago
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆849Updated this week
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,224Updated 3 months ago