Alternatives and similar repositories for DeepSpeed-Kernels

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

• IST-DASLab / Sparse-Marlin
  Boosting 4-bit inference kernels with 2:4 Sparsity
  ☆93Updated last year
• stanford-futuredata / stk
  ☆115Updated last year
• hpcaitech / TensorNVMe
  A Python library transfers PyTorch tensors between CPU and NVMe
  ☆125Updated last year
• cchan / tccl
  extensible collectives library in triton
  ☆95Updated 10 months ago
• IBM / triton-dejavu
  Framework to reduce autotune overhead to zero for well known deployments.
  ☆95Updated 4 months ago
• InternLM / turbomind
  ☆96Updated 10 months ago
• microsoft / AttentionEngine
  ☆117Updated 8 months ago
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆142Updated 8 months ago
• triton-lang / kernels
  ☆104Updated last year
• wangsiping97 / FastGEMV
  High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.
  ☆127Updated last year
• flashinfer-ai / cutlass-viz
  ☆65Updated 9 months ago
• tlc-pack / libflash_attn
  Standalone Flash Attention v2 kernel without libtorch dependency
  ☆113Updated last year
• usyd-fsalab / fp6_llm
  An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).
  ☆276Updated 6 months ago
• feifeibear / Odysseus-Transformer
  Odysseus: Playground of LLM Sequence Parallelism
  ☆79Updated last year
• anyscale / llm-continuous-batching-benchmarks
  ☆125Updated last year
• LeiWang1999 / AutoGPTQ.tvm
  GPTQ inference TVM kernel
  ☆40Updated last year
• PipeFusion / PipeFusion
  A Suite for Parallel Inference of Diffusion Transformers (DiTs) on multi-GPU Clusters
  ☆55Updated last year
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆73Updated 9 months ago
• SqueezeBits / QUICK
  QUICK: Quantization-aware Interleaving and Conflict-free Kernel for efficient LLM inference
  ☆120Updated last year
• foundation-model-stack / foundation-model-stack
  🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.
  ☆219Updated this week
• Infini-AI-Lab / MagicDec
  [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding
  ☆141Updated last year
• microsoft / TileFusion
  TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.
  ☆106Updated 7 months ago
• meta-pytorch / kraken
  Triton-based Symmetric Memory operators and examples
  ☆81Updated 3 weeks ago
• jundaf2 / INT8-Flash-Attention-FMHA-Quantization
  ☆160Updated 2 years ago
• NVIDIA / online-softmax
  Benchmark code for the "Online normalizer calculation for softmax" paper
  ☆105Updated 7 years ago
• meta-pytorch / applied-ai
  Applied AI experiments and examples for PyTorch
  ☆315Updated 5 months ago
• open-lm-engine / accelerated-model-architectures
  A bunch of kernels that might make stuff slower 😉
  ☆75Updated this week
• ankan-ban / llama_cu_awq
  llama INT4 cuda inference with AWQ
  ☆55Updated last year
• 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 7 months ago
• meta-pytorch / tritonbench
  Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.
  ☆324Updated this week