AMD 将在 PyTorchCon 2026 介绍 FlyDSL 集成 TorchInductor 方案
AMD 团队要演示怎么把自家 FlyDSL kernel DSL 接进 TorchInductor，还拿 Triton 跑分对比，做 GPU 优化的可以看看。
AMD 的 Liz Li 和 Jiahui Cao 将在 PyTorch Conference North America 2026 上讲解 FlyDSL，一个基于 Python 和 MLIR 的 GPU kernel DSL。该方案接入 TorchInductor 的编译与自动调优流水线，同时保留 torch.compile 的使用体验。演讲将展示 AMD Instinct GPU 上 transformer 训练与推理的性能结果，并对比 Triton 与 FlyDSL 两种实现。
Modern AI workloads rely on a growing range of specialized GPU kernels, and no single kernel generation backend is optimal across every operator.
At #PyTorchCon North America 2026, Liz Li and Jiahui Cao (@AMD) will present “Extending TorchInductor with FlyDSL: A New MLIR-Native Backend for High-Performance GEMMs.”
Liz and Jiahui will share how FlyDSL, AMD’s Python-native and MLIR-based GPU kernel DSL, integrates with TorchInductor’s compilation and autotuning pipeline while preserving the torch.compile user experience. The session will also present performance results for transformer training and inference on AMD Instinct GPUs, comparing Triton and FlyDSL implementations.
Register for PyTorch Conference North America 2026: https://t.co/jBApW8nESi
