Alternatives and similar repositories for triton

Users that are interested in triton are comparing it to the libraries listed below

• Dao-AILab / flash-attention
  Fast and memory-efficient exact attention
  ☆19,471Updated this week
• NVIDIA / FasterTransformer
  Transformer related optimization, including BERT, GPT
  ☆6,300Updated last year
• NVIDIA / TensorRT-LLM
  TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and support state-of-the-art optimizati…
  ☆11,576Updated this week
• NVIDIA / cutlass
  CUDA Templates for Linear Algebra Subroutines
  ☆8,427Updated last week
• huggingface / accelerate
  🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (i…
  ☆9,133Updated this week
• openxla / xla
  A machine learning compiler for GPUs, CPUs, and ML accelerators
  ☆3,505Updated this week
• NVIDIA / Megatron-LM
  Ongoing research training transformer models at scale
  ☆13,541Updated this week
• apache / tvm
  Open deep learning compiler stack for cpu, gpu and specialized accelerators
  ☆12,613Updated this week
• ggml-org / ggml
  Tensor library for machine learning
  ☆13,134Updated last week
• facebookresearch / xformers
  Hackable and optimized Transformers building blocks, supporting a composable construction.
  ☆9,930Updated this week
• triton-inference-server / server
  The Triton Inference Server provides an optimized cloud and edge inferencing solution.
  ☆9,755Updated last week
• facebookincubator / AITemplate
  AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (N…
  ☆4,675Updated 3 weeks ago
• bitsandbytes-foundation / bitsandbytes
  Accessible large language models via k-bit quantization for PyTorch.
  ☆7,584Updated this week
• deepspeedai / DeepSpeed
  DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
  ☆40,058Updated this week
• sgl-project / sglang
  SGLang is a fast serving framework for large language models and vision language models.
  ☆17,823Updated this week
• NVIDIA / TensorRT
  NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source compone…
  ☆12,144Updated last week
• flashinfer-ai / flashinfer
  FlashInfer: Kernel Library for LLM Serving
  ☆3,723Updated this week
• huggingface / safetensors
  Simple, safe way to store and distribute tensors
  ☆3,446Updated last week
• karpathy / llama2.c
  Inference Llama 2 in one file of pure C
  ☆18,735Updated last year
• NVIDIA / TransformerEngine
  A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Bla…
  ☆2,720Updated last week
• facebookresearch / fairscale
  PyTorch extensions for high performance and large scale training.
  ☆3,369Updated 4 months ago
• pytorch / torchtune
  PyTorch native post-training library
  ☆5,484Updated this week
• onnx / onnx
  Open standard for machine learning interoperability
  ☆19,582Updated last week
• pytorch / torchtitan
  A PyTorch native platform for training generative AI models
  ☆4,372Updated last week
• iree-org / iree
  A retargetable MLIR-based machine learning compiler and runtime toolkit.
  ☆3,347Updated this week
• karpathy / llm.c
  LLM training in simple, raw C/CUDA
  ☆27,588Updated 2 months ago
• NVIDIA / nccl
  Optimized primitives for collective multi-GPU communication
  ☆4,051Updated this week
• vllm-project / vllm
  A high-throughput and memory-efficient inference and serving engine for LLMs
  ☆58,035Updated this week
• meta-pytorch / gpt-fast
  Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
  ☆6,085Updated 3 weeks ago
• microsoft / LoRA
  Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"
  ☆12,660Updated 9 months ago