FlashSparse significantly reduces the computation redundancy for unstructured sparsity (for SpMM and SDDMM) on Tensor Cores through a Swap-and-Transpose mapping strategy. FlashSparse is accepted by PPoPP 2025.
☆39Oct 5, 2025Updated 4 months ago
Alternatives and similar repositories for FlashSparse
Users that are interested in FlashSparse are comparing it to the libraries listed below
Sorting:
- Official implementation of Acc-SpMM: Accelerating General-purpose Sparse Matrix-Matrix Multiplication with GPU Tensor Cores.☆14Nov 13, 2025Updated 3 months ago
- Magicube is a high-performance library for quantized sparse matrix operations (SpMM and SDDMM) of deep learning on Tensor Cores.☆91Nov 23, 2022Updated 3 years ago
- ☆10Mar 2, 2024Updated last year
- ☆14Apr 24, 2024Updated last year
- Aligntune : A Modular Toolkit for Post Training Alignment of LLMs☆33Updated this week
- Artifact for USENIX ATC'23: TC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs.☆53Oct 16, 2023Updated 2 years ago
- ☆20Aug 21, 2023Updated 2 years ago
- ☆46Jun 19, 2024Updated last year
- Fast and low-memory attention layer written in CUDA☆20Jul 14, 2023Updated 2 years ago
- ☆27Mar 24, 2025Updated 11 months ago
- ☆17Feb 26, 2020Updated 6 years ago
- ☆31Apr 2, 2025Updated 10 months ago
- Source code of the SC '23 paper: "DASP: Specific Dense Matrix Multiply-Accumulate Units Accelerated General Sparse Matrix-Vector Multipli…☆28Jun 18, 2024Updated last year
- ☆48Aug 23, 2021Updated 4 years ago
- ☆42Nov 1, 2025Updated 3 months ago
- Horizontal Fusion☆24Jan 7, 2022Updated 4 years ago
- Sparse linear Boolean algebra for Nvidia Cuda☆26Nov 17, 2025Updated 3 months ago
- PyGim is the first runtime framework to efficiently execute Graph Neural Networks (GNNs) on real Processing-in-Memory systems. It provide…☆33Apr 23, 2025Updated 10 months ago
- A SapientML plugin of SapientMLGenerator☆11Dec 23, 2025Updated 2 months ago
- A library of GPU kernels for sparse matrix operations.☆283Nov 24, 2020Updated 5 years ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆70Mar 20, 2025Updated 11 months ago
- ☆32Aug 24, 2022Updated 3 years ago
- ☆35Apr 10, 2024Updated last year
- a size profiler for cuda binary☆72Jan 15, 2026Updated last month
- EdgeCortix maintained and extended fork of Apache TVM compiler stack utilized by MERA framework. TVM is an open deep learning compiler st…☆11Dec 22, 2023Updated 2 years ago
- AI-ML-NLP Task Group☆13Aug 10, 2023Updated 2 years ago
- Continuous Pipelined Speculative Decoding☆16Jan 4, 2026Updated last month
- ☆10Mar 8, 2025Updated 11 months ago
- Notes and Examples to get started Parallel Computing with CUDA.☆13Nov 1, 2019Updated 6 years ago
- SIMPLER MAGIC: Synthesis and In-memory MaPping of Logic Execution in a single Row for Memristor Aided loGIC☆12Dec 5, 2019Updated 6 years ago
- Synthesis using Synopsys DC and Physical Design flow using Synopsys ICC II, of my RISC-V 5 stage pipelined using 32 nm tech repo☆14Jul 31, 2024Updated last year
- Explains the conclusions of a logic program.☆10May 25, 2023Updated 2 years ago
- A graph linear algebra overlay☆51Apr 26, 2023Updated 2 years ago
- ☆10May 24, 2021Updated 4 years ago
- Residual vector quantization for KV cache compression in large language model☆11Oct 22, 2024Updated last year
- RISC-V kernel step-by-step implmenetation☆12Nov 12, 2019Updated 6 years ago
- A DL compiler fuzzer☆14Nov 1, 2024Updated last year
- FPGA-based HyperLogLog Accelerator☆12Jul 13, 2020Updated 5 years ago
- ☆23Dec 30, 2025Updated 2 months ago