Keynote talk delivered by Didem Unat at Intl Conference on Performance Engineering 2026

KaMPIng: (Near) zero-overhead MPI wrapper for modern C++

Simple message passing library

BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.

Official inference framework for 1-bit LLMs

PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for evaluation of training and inference platforms.

📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉

Blue Gene/Q driver, see https://repo.anl-external.org/repos/bgq-driver/

🍦 Never use cout/printf to debug again

A Fortran linter, written in Rust and installable with Python.

Parallel Computing -- Validation Suite: Validation engine for Exascale project benchmarks

Fortran bindings generated using Coccinelle

Modern C++ Programming Course (C++03/11/14/17/20/23/26)

Test if AVX vector loads and stores are atomic

A place to store information for the tensor discussions and possible specifications.

do a simple closed shell Hartree-Fock using McMurchie-Davidson to compute integrals

Data structures, algorithms, and C++ reference library

LLM inference in Fortran

Efficient numerical computation of the Pfaffian for dense and banded skew-symmetric matrices

A framework and suite of cases for testing a Fortran compiler

Fast CUDA matrix multiplication from scratch

Flexible and performant GEMM kernels in Julia

Geometry optimization code that includes the TRIC coordinate system

optking: A molecular geometry optimization program

Molecular structure optimizer

Repository to host supporting information and code samples for Accelerated DFT

LLM training in simple, raw C/CUDA