Eikan Wang

Accelerator and Quantization Support on ExecuTorch - Chen Lai, Meta Agents in Space - George Williams, Armijo Innovations Enabling End-to-end Embedded ML Applications With ExecuTorch for the Arm Ethos-U NPU - George Gekov, Arm ExecuTorch for ML Inference on Edge Devices Running Zephyr RTOS - Bujji...

Beyond One-Size-Fits-All: Building Domain-Specific Language Models With PyTorch - Mayank Kumar, DeepTempo Double LLM RAG Accuracy by Adding GNNs - Serge Panev, NVIDIA Serving Transformers at Scale: KV-Cache-Aware Routing for PyTorch Inference - Maroon Ayoub, IBM & Tyler Michael Smith, Red Hat...

A Case Study of Torch.compile-induced Timeouts in Distributed - Simon Fan & Brian Hirsh, Meta Activation Checkpoint 2: A New Graph-capture-based Activation Checkpoint - Jeffrey Wan, Meta Automatic Parallelism in PyTorch - Will Constable & Francisco Massa, Meta Breaking Less: Tackling Graph Breaks...

PyTorch has become a cornerstone for AI development, powering everything from inference pipelines to autonomous agents. Traditionally, these workloads required high-end systems equipped with discrete GPUs, often limiting accessibility and flexibility. Scaling inference at this pace demands more...

Purge the GIL: Improved Torch.DataLoader - Michal Szolucha & Rostan Tabet, NVIDIA XFormers - Daniel Haziza, Meta AI  TritonCC: AOT Triton Workflow for TorchScript C++ Runtime - Sijia Chen & Huamin Li, Meta The PyTorch 2.0 Inference Story - Angela Yi, Bin Bao, Sheng Qin & Sherlock Huang, Meta Tensor...

- There are two integration levels to add a new backend for the PyTorch compiler - AtenIR/PrimsIR level and Inductor loop IR level. The ATen/Prim level IR integration has been there via the custom backend registration infrastructure (https://pytorch.org/docs/stable/dynamo/custom-backends.html)....