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.
☆2,720Updated 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☆3,723Updated this week
- A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. …☆1,344Updated this week
- PyTorch native quantization and sparsity for training and inference☆2,341Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,496Updated last year
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,246Updated last month
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆2,159Updated last month
- Transformer related optimization, including BERT, GPT☆6,300Updated last year
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,053Updated 2 months ago
- Pipeline Parallelism for PyTorch☆779Updated last year
- PyTorch extensions for high performance and large scale training.☆3,369Updated 4 months ago
- Tile primitives for speedy kernels☆2,688Updated this week
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,829Updated 2 weeks ago
- Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".☆2,179Updated last year
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆1,608Updated this 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…☆1,583Updated last year
- A Python-level JIT compiler designed to make unmodified PyTorch programs faster.☆1,062Updated last year
- SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX R…☆2,491Updated this week
- The Triton TensorRT-LLM Backend☆887Updated last week
- A pytorch quantization backend for optimum☆985Updated 2 weeks ago
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,247Updated 4 months ago
- Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 using FP8/NVFP4/MXFP4☆913Updated this week
- Microsoft Automatic Mixed Precision Library☆619Updated 11 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆895Updated last year
- Minimalistic large language model 3D-parallelism training☆2,191Updated last week
- Distributed Compiler based on Triton for Parallel Systems☆1,107Updated this week
- FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/☆1,440Updated this week
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,276Updated 6 months ago
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆1,415Updated last year
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆1,773Updated this week
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆863Updated this week