pytorch / ao
PyTorch native quantization and sparsity for training and inference
☆2,015Updated this week
Alternatives and similar repositories for ao:
Users that are interested in ao are comparing it to the libraries listed below
- A PyTorch native library for large-scale model training☆3,665Updated this week
- Thunder gives you PyTorch models superpowers for training and inference. Unlock out-of-the-box optimizations for performance, memory and …☆1,335Updated this week
- FlashInfer: Kernel Library for LLM Serving☆2,764Updated this week
- Minimalistic large language model 3D-parallelism training☆1,836Updated this week
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Bla…☆2,390Updated this week
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆2,970Updated 3 weeks ago
- Tile primitives for speedy kernels☆2,312Updated this week
- A pytorch quantization backend for optimum☆927Updated last week
- nvidia-modelopt is a unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculat…☆896Updated this week
- Puzzles for learning Triton☆1,603Updated 5 months ago
- Official implementation of Half-Quadratic Quantization (HQQ)☆800Updated this week
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆1,301Updated this week
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,397Updated 9 months ago
- Minimalistic 4D-parallelism distributed training framework for education purpose☆1,346Updated last month
- 🚀 Efficient implementations of state-of-the-art linear attention models in Torch and Triton☆2,344Updated this week
- Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA☆820Updated this week
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆536Updated this week
- PyTorch native post-training library☆5,138Updated this week
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,136Updated last week
- [ICLR 2025] Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling☆867Updated this week
- GPU programming related news and material links☆1,480Updated 4 months ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆1,089Updated this week
- Pipeline Parallelism for PyTorch☆765Updated 8 months ago
- A throughput-oriented high-performance serving framework for LLMs☆804Updated this week
- SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX R…☆2,387Updated this week
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆810Updated 8 months ago
- GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection☆1,546Updated 6 months ago
- Helpful tools and examples for working with flex-attention☆746Updated 3 weeks ago
- On-device AI across mobile, embedded and edge for PyTorch☆2,790Updated this week
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,007Updated last month