tsinghua-ideal / Canvas
Canvas: End-to-End Kernel Architecture Search in Neural Networks
☆26Updated 5 months ago
Alternatives and similar repositories for Canvas:
Users that are interested in Canvas are comparing it to the libraries listed below
- Tacker: Tensor-CUDA Core Kernel Fusion for Improving the GPU Utilization while Ensuring QoS☆25Updated 2 months ago
- ThrillerFlow is a Dataflow Analysis and Codegen Framework written in Rust.☆14Updated 5 months ago
- MAGIS: Memory Optimization via Coordinated Graph Transformation and Scheduling for DNN (ASPLOS'24)☆50Updated 11 months ago
- An extention of TVMScript to write simple and high performance GPU kernels with tensorcore.☆50Updated 9 months ago
- ☆39Updated 11 months ago
- Supplemental materials for The ASPLOS 2025 / EuroSys 2025 Contest on Intra-Operator Parallelism for Distributed Deep Learning☆23Updated 4 months ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆44Updated last month
- ☆19Updated 7 months ago
- Artifacts of EVT ASPLOS'24☆24Updated last year
- ASPLOS'24: Optimal Kernel Orchestration for Tensor Programs with Korch☆34Updated last month
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆34Updated 2 weeks ago
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆37Updated 4 months ago
- ☆37Updated 9 months ago
- ☆79Updated 2 years ago
- Triton to TVM transpiler.☆19Updated 6 months ago
- ☆22Updated 2 years ago
- PerFlow-AI is a programmable performance analysis, modeling, prediction tool for AI system.☆19Updated last week
- LLM Inference analyzer for different hardware platforms☆65Updated last week
- ☆28Updated 9 months ago
- ☆9Updated last year
- play gemm with tvm☆91Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆84Updated this week
- Repository for artifact evaluation of ASPLOS 2023 paper "SparseTIR: Composable Abstractions for Sparse Compilation in Deep Learning"☆25Updated 2 years ago
- ☆105Updated last week
- ☆60Updated last month
- ☆57Updated last week
- DietCode Code Release☆63Updated 2 years ago
- Magicube is a high-performance library for quantized sparse matrix operations (SpMM and SDDMM) of deep learning on Tensor Cores.☆87Updated 2 years ago
- Automatic Mapping Generation, Verification, and Exploration for ISA-based Spatial Accelerators☆108Updated 2 years ago
- Stateful LLM Serving☆65Updated last month