☆71Mar 26, 2025Updated 11 months ago
Alternatives and similar repositories for DeepSpeed-Kernels
Users that are interested in DeepSpeed-Kernels are comparing it to the libraries listed below
Sorting:
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆96Feb 20, 2026Updated last week
- ☆13Jan 7, 2025Updated last year
- Persistent dense gemm for Hopper in `CuTeDSL`☆15Aug 9, 2025Updated 6 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆106Jun 28, 2025Updated 8 months ago
- ☆261Jul 11, 2024Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆128Jul 13, 2024Updated last year
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,095Jun 30, 2025Updated 8 months ago
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆46Jun 11, 2025Updated 8 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆44Feb 27, 2025Updated last year
- Implementation of IceFormer: Accelerated Inference with Long-Sequence Transformers on CPUs (ICLR 2024).☆25Updated this week
- A Triton JIT runtime and ffi provider in C++☆31Updated this week
- ☆60Updated this week
- Perplexity GPU Kernels☆564Nov 7, 2025Updated 3 months ago
- Official resporitory for "IPDPS' 24 QSync: Quantization-Minimized Synchronous Distributed Training Across Hybrid Devices".☆20Feb 23, 2024Updated 2 years ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,018Sep 4, 2024Updated last year
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆469Feb 21, 2026Updated last week
- Dynamic Memory Management for Serving LLMs without PagedAttention☆463May 30, 2025Updated 8 months ago
- ☆20Sep 28, 2024Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency☆114Sep 10, 2024Updated last year
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆336Jul 2, 2024Updated last year
- extensible collectives library in triton☆95Mar 31, 2025Updated 11 months ago
- CUDA 12.2 HMM demos☆20Jul 26, 2024Updated last year
- IntLLaMA: A fast and light quantization solution for LLaMA☆18Jul 21, 2023Updated 2 years ago
- ☆65Apr 26, 2025Updated 10 months ago
- ☆131Nov 11, 2024Updated last year
- Torch Distributed Experimental☆117Aug 5, 2024Updated last year
- optimized BERT transformer inference on NVIDIA GPU. https://arxiv.org/abs/2210.03052☆478Mar 15, 2024Updated last year
- Sample Codes using NVSHMEM on Multi-GPU☆30Jan 22, 2023Updated 3 years ago
- GPTQ inference Triton kernel☆321May 18, 2023Updated 2 years ago
- A low-latency & high-throughput serving engine for LLMs☆480Jan 8, 2026Updated last month
- Stateful LLM Serving☆96Mar 11, 2025Updated 11 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆72Sep 8, 2024Updated last year
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆277Jul 16, 2025Updated 7 months ago
- ☆104Sep 9, 2024Updated last year
- ☆118May 19, 2025Updated 9 months ago
- ☆24Feb 20, 2024Updated 2 years ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆234Sep 24, 2023Updated 2 years ago
- Distributed SDDMM Kernel☆12Jul 8, 2022Updated 3 years ago
- CPU and GPU tutorial examples☆13Apr 4, 2025Updated 10 months ago