high performance computing on graphics processing units: hgpu.org

Jan-Frederik Schulte, Benjamin Ramhorst, Chang Sun, Jovan Mitrevski, Nicolò Ghielmetti, Enrico Lupi, Dimitrios Danopoulos, Vladimir Loncar, Javier Duarte, David Burnette, Lauri Laatu, Stylianos Tzelepis, Konstantinos Axiotis, Quentin Berthet, Haoyan Wang, Paul White, Suleyman Demirsoy, Marco Colombo, Thea Aarrestad, Sioni Summers, Maurizio Pierini, Giuseppe Di Guglielmo, Jennifer Ngadiuba, Javier Campos, Ben Hawks, Abhijith Gandrakota, Farah Fahim, Nhan Tran, George Constantinides, Zhiqiang Que, Wayne Luk, Alexander Tapper, Duc Hoang, Noah Paladino, Philip Harris, Bo-Cheng Lai, Manuel Valentin, Ryan Forelli, Seda Ogrenci, Lino Gerlach, Rian Flynn, Mia Liu, Daniel Diaz, Elham Khoda, Melissa Quinnan, Russell Solares, Santosh Parajuli, Mark Neubauer, Christian Herwig, Ho Fung Tsoi, Dylan Rankin, Shih-Chieh Hsu, Scott Hauck

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, FPGA, HLS, Machine learning, Package

December 7, 2025 by hgpu

Decoupled Triton: A Block-Level Decoupled Language for Writing and Exploring Efficient Machine-Learning Kernels

Quinn Leo Pham

View

Download (PDF)

Tags: Compilers, Computer science, Machine learning, nVidia, nVidia RTX 5000 Ada, PyTorch, Thesis, Triton

December 7, 2025 by hgpu

high performance computing on graphics processing units: hgpu.org

CuTeGen: An LLM-Based Agentic Framework for Generation and Optimization of High-Performance GPU Kernels using CuTe

AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search

Making LLMs Optimize Multi-Scenario CUDA Kernels Like Experts

SynPerf: A Hybrid Analytical-ML Framework for GPU Performance Prediction

Equivalence Checking of ML GPU Kernels

AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization

Memory-Efficient Acceleration of Block Low-Rank Foundation Models on Resource Constrained GPUs

ML Inference Scheduling with Predictable Latency

TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization

Microbenchmarking NVIDIA’s Blackwell Architecture: An in-depth Architectural Analysis

hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware

Decoupled Triton: A Block-Level Decoupled Language for Writing and Exploring Efficient Machine-Learning Kernels

Recent source codes

MegaTrain: Full Precision Training of 100B+ Parameter Large Language Models on a Single GPU

Device Virtual Machine (DVM)

Agentic Code Optimization via Compiler-LLM Cooperation

AutoKernel: Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels

Triton-Sanitizer: A Fast and Device-Agnostic Memory Sanitizer for Triton with Rich Diagnostic Context

LLM.Q: Quantized LLM training in pure CUDA/C++

SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits

True 4-Bit Quantized CNN Training on CPU

cuFuzz: A GPU-oriented coverage-guided fuzzer for userland CUDA application

KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization

Most viewed papers (last 30 days)