# Glow: Graph Lowering Compiler Techniques for Neural Networks

arXiv: https://arxiv.org/abs/1805.00907 · HTML: https://mlsanity.com/paper/1805.00907/

## Key facts

- Cited by 75 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 May 2018.

## Built on by (most cited first, 30 of 75)

- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Pruning and Quantization for Deep Neural Network Acceleration: A Survey](https://mlsanity.com/paper/2101.09671/) (2021)
- [Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better](https://mlsanity.com/paper/2106.08962/) (2021)
- [GSPMD: General and Scalable Parallelization for ML Computation Graphs](https://mlsanity.com/paper/2105.04663/) (2021)
- [Full Stack Optimization of Transformer Inference: a Survey](https://mlsanity.com/paper/2302.14017/) (2023)
- [Astra: A Multi-Agent System for GPU Kernel Performance Optimization](https://mlsanity.com/paper/2509.07506/) (2025)
- [ThunderKittens: Simple, Fast, and Adorable AI Kernels](https://mlsanity.com/paper/2410.20399/) (2024)
- [Lancet: Accelerating Mixture-of-Experts Training via Whole Graph Computation-Communication Overlapping](https://mlsanity.com/paper/2404.19429/) (2024)
- [On-device Training: A First Overview on Existing Systems](https://mlsanity.com/paper/2212.00824/) (2022)
- [Learning to Reverse DNNs from AI Programs Automatically](https://mlsanity.com/paper/2205.10364/) (2022)
- [LazyTensor: combining eager execution with domain-specific compilers](https://mlsanity.com/paper/2102.13267/) (2021)
- [Revisiting Neuron Coverage for DNN Testing: A Layer-Wise and Distribution-Aware Criterion](https://mlsanity.com/paper/2112.01955/) (2021)
- [UNIT: Unifying Tensorized Instruction Compilation](https://mlsanity.com/paper/2101.08458/) (2021)
- [Coverage-Guided Tensor Compiler Fuzzing with Joint IR-Pass Mutation](https://mlsanity.com/paper/2202.09947/) (2022)
- [MetaTune: Meta-Learning Based Cost Model for Fast and Efficient Auto-tuning Frameworks](https://mlsanity.com/paper/2102.04199/) (2021)
- [Horizontally Fused Training Array: An Effective Hardware Utilization Squeezer for Training Novel Deep Learning Models](https://mlsanity.com/paper/2102.02344/) (2021)
- [Federated Remote Physiological Measurement with Imperfect Data](https://mlsanity.com/paper/2203.05759/) (2022)
- [Decompiling x86 Deep Neural Network Executables](https://mlsanity.com/paper/2210.01075/) (2022)
- [Demand Layering for Real-Time DNN Inference with Minimized Memory Usage](https://mlsanity.com/paper/2210.04024/) (2022)
- [Swift for TensorFlow: A portable, flexible platform for deep learning](https://mlsanity.com/paper/2102.13243/) (2021)
- [Quantune: Post-training Quantization of Convolutional Neural Networks using Extreme Gradient Boosting for Fast Deployment](https://mlsanity.com/paper/2202.05048/) (2022)
- [Special Session: Towards an Agile Design Methodology for Efficient, Reliable, and Secure ML Systems](https://mlsanity.com/paper/2204.09514/) (2022)
- [Bring Your Own Codegen to Deep Learning Compiler](https://mlsanity.com/paper/2105.03215/) (2021)
- [Deep Learning Library Testing: Definition, Methods and Challenges](https://mlsanity.com/paper/2404.17871/) (2024)
- [From Tiny Machine Learning to Tiny Deep Learning: A Survey](https://mlsanity.com/paper/2506.18927/) (2025)
- [NullaNet Tiny: Ultra-low-latency DNN Inference Through Fixed-function Combinational Logic](https://mlsanity.com/paper/2104.05421/) (2021)
- [Memory Safe Computations with XLA Compiler](https://mlsanity.com/paper/2206.14148/) (2022)
- [MSRL: Distributed Reinforcement Learning with Dataflow Fragments](https://mlsanity.com/paper/2210.00882/) (2022)
- [A Data-Centric Optimization Framework for Machine Learning](https://mlsanity.com/paper/2110.10802/) (2021)
- [Benchmarking the Linear Algebra Awareness of TensorFlow and PyTorch](https://mlsanity.com/paper/2202.09888/) (2022)

## Abstract (arXiv)

This paper presents the design of Glow, a machine learning compiler for heterogeneous hardware. It is a pragmatic approach to compilation that enables the generation of highly optimized code for multiple targets. Glow lowers the traditional neural network dataflow graph into a two-phase strongly-typed intermediate representation. The high-level intermediate representation allows the optimizer to perform domain-specific optimizations. The lower-level instruction-based address-only intermediate representation allows the compiler to perform memory-related optimizations, such as instruction scheduling, static memory allocation and copy elimination. At the lowest level, the optimizer performs machine-specific code generation to take advantage of specialized hardware features. Glow features a lowering phase which enables the compiler to support a high number of input operators as well as a large number of hardware targets by eliminating the need to implement all operators on all targets. The lowering phase is designed to reduce the input space and allow new hardware backends to focus on a small number of linear algebra primitives.