MAX’s Graph Compiler Internals with Feras Boulala
Автор: Modular
Загружено: 2026-01-23
Просмотров: 501
In this talk, Modular's Feras Boulala dives into MAX's Graph Compiler, showing how it transforms deep learning graphs into efficient, high-performance Mojo code. He explains how custom ops, structured kernels, and fusion techniques enable flexibility, scalability, and productivity.
Learn how Modular achieves hardware-agnostic performance, making it easier to optimize and run AI models across NVIDIA, AMD, and other platforms while keeping code readable, extensible, and maintainable.
00:00 Introduction & What is a Graph Compiler?
03:45 Three Compiler Phases (RMO/MO, MOG, MGP)
06:40 The Kernel vs Compiler Spectrum
08:30 Modular's Solution: Extensibility First
11:30 Bridging Mojo to MLIR & Fusion
16:00 Future Roadmap
17:00 Q&A Session
Learn more about MAX: https://www.modular.com/max
Join our community 🤝:
Forum - https://forum.modular.com/
GitHub - https://github.com/modular
X (aka Twitter) - https://x.com/modular
LinkedIn - / modular-ai
Reddit - / modularai
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
![Как происходит модернизация остаточных соединений [mHC]](https://imager.clipsaver.ru/jYn_1PpRzxI/max.jpg)
