AMD to present FlyDSL, an MLIR-native GPU kernel backend for TorchInductor, at PyTorchCon
PyTorch · x · 2026-10-03
PyTorch announced that at PyTorch Conference North America 2026, AMD's Liz Li and Jiahui Cao will present FlyDSL, AMD's Python-native, MLIR-based GPU kernel DSL.
- Modern AI workloads rely on specialized GPU kernels, and no single kernel generation backend is optimal across every operator.
- FlyDSL integrates with TorchInductor's compilation and autotuning pipeline while preserving the torch.compile user experience.
- The session includes performance results for transformer training and inference on AMD Instinct GPUs, comparing Triton vs. FlyDSL implementations.
More from Infra
- Runner pays $4,500 for a third GPU to run near-frontier models locally — marian_nmt · 2026-10-03
- 90% of AI chip value accrues to incumbents, no OpenAI equivalent for years — menhguin · 2026-10-03
- MrKan ships Docker edition for air-gapped, offline AI agent kanban boards — agf2007 · 2026-10-03
- SK Hynix's Solidigm Reportedly Weighing $150 Billion IPO, Nearly Triple Arm's — Beth_Kindig · 2026-10-03
- ezyang's DeepSeek-V3 series part 3: roofline analysis for training DSv3 on Hopper — ezyang · 2026-10-03
- Developer slams DeepInfra's 33% input token price hike with just 2 days notice — julianharris · 2026-10-03