about
Architecting for HW Diversity: Meta's E2E Co-Design for AI Training on AMD GPUs (arxiv.org)
3 points by matt_d 30 days ago | hide | past | pdf | discuss on HN

In plain words: Instead of copying NVIDIA-tuned code, Meta reshaped models and software to fit AMD chips' strengths, with a predictor that picks the best routines without test runs. AI coding tools that rewrite NVIDIA code made training up to 1.5 times faster than a straight port.

Abstract · Architecting for Hardware Diversity: Meta's End-to-End Co-Design for High-Performance AI Training on AMD GPUs

Meta deploys a heterogeneous fleet comprising NVIDIA GPUs, AMD GPUs, and proprietary MTIA accelerators to reduce costs and mitigate vendor lock-in. However, the extensive history of NVIDIA-specific optimizations hinders performance portability to other hardware. This paper presents our methods for optimizing alternative accelerators, using AMD Instinct MI350X versus NVIDIA B200 for recommendation model training as a case study. We propose three reusable principles. First, hardware-model co-design is essential: rather than directly porting models from NVIDIA GPUs, revising model structures and software recipes allows full exploitation of accelerator-specific strengths (e.g., MI350X's 1.6x HBM capacity relative to B200). Second, our analytical performance model enables zero-warmup optimal kernel selection across diverse datatypes and matrix shapes in production models, drastically simplifying tuning. Third, increasingly mature agentic workflows have lowered the barrier to porting from the CUDA ecosystem; we extend these workflows for microarchitecture-aware kernel optimizations and custom shim generation. Compared to the directly ported baseline, these methods deliver up to 1.5x improvement in end-to-end training throughput. To our knowledge, this is the first hyperscaler report on productionizing AMD GPUs for large-scale training. Our contributions have been upstreamed to the open-source PyTorch, Triton, and ROCm ecosystems.

Haoyu Zhang, Arshad Rahman, Adnan Aziz, Ajith Prabhudev, Amin Farmahini, Ben Rhoads, Benson Ma, Brian Shi, Cen Zhao, CK Luk, Dingming Wu, Emad El-Haraty, et al.
arXiv:2610.22230 · cs.DC, cs.AR, cs.LG · submitted Oct 15, 2026
abstract · pdf · html · 16 pages, 21 figures, 6 tables. To appear in the 59th IEEE/ACM International Symposium on Microarchitecture (MICRO 2026)

add comment on HN