Alternatives and similar repositories for gpu_kernels

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

• IST-DASLab / Sparse-Marlin
  Boosting 4-bit inference kernels with 2:4 Sparsity
  ☆85Updated last year
• SqueezeBits / QUICK
  QUICK: Quantization-aware Interleaving and Conflict-free Kernel for efficient LLM inference
  ☆118Updated last year
• meta-pytorch / KernelAgent
  Autonomous GPU Kernel Generation via Deep Agents
  ☆137Updated this week
• IBM / triton-dejavu
  Framework to reduce autotune overhead to zero for well known deployments.
  ☆85Updated 2 months ago
• microsoft / AttentionEngine
  ☆109Updated 6 months ago
• meta-pytorch / kraken
  Triton-based Symmetric Memory operators and examples
  ☆63Updated last month
• stanford-futuredata / stk
  ☆113Updated last year
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆130Updated 5 months ago
• ankan-ban / llama_cu_awq
  llama INT4 cuda inference with AWQ
  ☆55Updated 10 months ago
• tile-ai / AttentionEngine
  ☆50Updated 6 months ago
• deepspeedai / DeepSpeed-Kernels
  ☆71Updated 7 months ago
• cchan / tccl
  extensible collectives library in triton
  ☆91Updated 7 months ago
• triton-lang / kernels
  ☆93Updated last year
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆73Updated 6 months ago
• INT-FlashAttention2024 / INT-FlashAttention
  ☆83Updated 9 months ago
• efeslab / fiddler
  [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration
  ☆243Updated last year
• wangsiping97 / FastGEMV
  High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.
  ☆122Updated last year
• Dao-AILab / grouped-latent-attention
  ☆130Updated 5 months ago
• microsoft / TileFusion
  TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.
  ☆100Updated 4 months ago
• LeiWang1999 / AutoGPTQ.tvm
  GPTQ inference TVM kernel
  ☆39Updated last year
• tlc-pack / libflash_attn
  Standalone Flash Attention v2 kernel without libtorch dependency
  ☆112Updated 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.
  ☆45Updated 5 months ago
• flashinfer-ai / cutlass-viz
  ☆65Updated 6 months ago
• opengear-project / GEAR
  GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM
  ☆170Updated last year
• li-plus / flash-preference
  Accelerate LLM preference tuning via prefix sharing with a single line of code
  ☆51Updated 4 months ago
• IST-DASLab / qutlass
  QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning
  ☆134Updated last week
• gpu-mode / ring-attention
  ring-attention experiments
  ☆155Updated last year
• feifeibear / ChituAttention
  Quantized Attention on GPU
  ☆44Updated last year
• Infini-AI-Lab / MagicDec
  [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding
  ☆131Updated 11 months ago
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆223Updated 2 years ago