ENERZAi·Oct 20Optimizing the MLIR-Based Torq NPU CompilerHi, this is ENERZAi. We recently contributed to the backend implementation and model optimization of the Torq compiler, an open-source…
ENERZAi·Oct 20MLIR 기반 Torq NPU 컴파일러 개발 여정안녕하세요, 에너자이입니다. 에너자이는 최근 글로벌 반도체사인 Synaptics의 Astra SL2610 제품군에 탑재된 Torq NPU를 위해, 오픈소스 IREE/MLIR 기반 Torq 컴파일러의 백엔드 구현과 모델 최적화에 참여했는데요. 오늘은…
InQuansightbyQuansight·Sep 18Finch meets MLIRThis blog was first published on the Quansight Blog by Yu Sheng Aow.
smolnero·Aug 24LLVM for the Rest of UsLLVM is everywhere. It sits underneath an enormous amount of modern software and touches everything from ordinary CPU compilation to GPU…
Jialei Chen·Jul 4When Execution Strategy Becomes Program Meaning: Separation of Concerns in GPU CompilersIntroduction
Botsz·Apr 1MLIR: My Learning Journey — InstallationDisclaimer: This is a documentation of my learning process only. Following these steps does not guarantee identical results.A response icon1A response icon1
Ayoub Ben Chaliah·Feb 19TensaLang: A programming language for LLM inference | Making optimized inference engines easyTensaLang is a language + compiler + runtime for writing LLM forward passes and sampling in source code, lowering through MLIR to CPU or…A response icon5A response icon5
Alexander Novikov·Feb 5MLIR, Nanopass, and Green-Red Trees: Compiler Ideas That Make Code Generators BetterMost code generators are monolithic. A single class reads some input, makes a bunch of decisions, and spits out source files. It works…
Alex Wang·Jan 30Two Small Experiments to Understand “Lowering” in MLIRThis note records two quick experiments I used to make the idea of “Lowering” concrete.
Samarth Narang·Dec 9, 2025Triton on NPUs: What Changes When You Leave the GPU WorldA systems-oriented walkthrough of what changes when you bring Triton’s programming model from GPUs to NPUs: execution model, memory…
