Alternatives and similar repositories for DeepSpeed-Kernels

Users that are interested in DeepSpeed-Kernels are comparing it to the libraries listed below

• hpcaitech / TensorNVMe
  A Python library transfers PyTorch tensors between CPU and NVMe
  ☆123Updated last year
• IBM / triton-dejavu
  Framework to reduce autotune overhead to zero for well known deployments.
  ☆90Updated 3 months ago
• stanford-futuredata / stk
  ☆115Updated last year
• cchan / tccl
  extensible collectives library in triton
  ☆91Updated 8 months ago
• IST-DASLab / Sparse-Marlin
  Boosting 4-bit inference kernels with 2:4 Sparsity
  ☆89Updated last year
• microsoft / AttentionEngine
  ☆114Updated 7 months ago
• InternLM / turbomind
  ☆97Updated 8 months ago
• anyscale / llm-continuous-batching-benchmarks
  ☆122Updated last year
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆135Updated 6 months ago
• usyd-fsalab / fp6_llm
  An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).
  ☆276Updated 5 months ago
• triton-lang / kernels
  ☆99Updated last year
• wangsiping97 / FastGEMV
  High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.
  ☆123Updated last year
• microsoft / TileFusion
  TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.
  ☆105Updated 5 months ago
• feifeibear / Odysseus-Transformer
  Odysseus: Playground of LLM Sequence Parallelism
  ☆78Updated last year
• foundation-model-stack / foundation-model-stack
  🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.
  ☆216Updated last week
• meta-pytorch / kraken
  Triton-based Symmetric Memory operators and examples
  ☆67Updated 2 months ago
• LeiWang1999 / AutoGPTQ.tvm
  GPTQ inference TVM kernel
  ☆41Updated last year
• ankan-ban / llama_cu_awq
  llama INT4 cuda inference with AWQ
  ☆55Updated 11 months ago
• Bruce-Lee-LY / decoding_attention
  Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.
  ☆46Updated 6 months ago
• meta-pytorch / applied-ai
  Applied AI experiments and examples for PyTorch
  ☆311Updated 4 months ago
• flashinfer-ai / cutlass-viz
  ☆65Updated 7 months ago
• tlc-pack / libflash_attn
  Standalone Flash Attention v2 kernel without libtorch dependency
  ☆112Updated last year
• gpu-mode / ring-attention
  ring-attention experiments
  ☆160Updated last year
• Infini-AI-Lab / MagicDec
  [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding
  ☆135Updated last year
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆73Updated 7 months ago
• NVIDIA / online-softmax
  Benchmark code for the "Online normalizer calculation for softmax" paper
  ☆103Updated 7 years ago
• fw-ai / llama-cuda-graph-example
  Example of applying CUDA graphs to LLaMA-v2
  ☆12Updated 2 years ago
• AlibabaPAI / FLASHNN
  ☆103Updated last year
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆230Updated 2 years ago
• neuralmagic / AutoFP8
  ☆206Updated 7 months ago