# QONNX: Representing Arbitrary-Precision Quantized Neural Networks

arXiv: https://arxiv.org/abs/2206.07527 · HTML: https://mlsanity.com/paper/2206.07527/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 29 references that are in the map.
- First posted on arXiv 15 Jun 2022.

## Builds on (most cited first, 15 of 15)

- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [BinaryConnect: Training Deep Neural Networks with binary weights during propagations](https://mlsanity.com/paper/1511.00363/) (2015)
- [HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision](https://mlsanity.com/paper/1905.03696/) (2019)
- [HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networks](https://mlsanity.com/paper/1911.03852/) (2019)
- [Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weights](https://mlsanity.com/paper/1702.03044/) (2017)
- [FINN: A Framework for Fast, Scalable Binarized Neural Network Inference](https://mlsanity.com/paper/1612.07119/) (2016)
- [Fast inference of deep neural networks in FPGAs for particle physics](https://mlsanity.com/paper/1804.06913/) (2018)
- [Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectors](https://mlsanity.com/paper/2006.10159/) (2020)
- [FINN-R: An End-to-End Deep-Learning Framework for Fast Exploration of Quantized Neural Networks](https://mlsanity.com/paper/1809.04570/) (2018)
- [Efficient Execution of Quantized Deep Learning Models: A Compiler Approach](https://mlsanity.com/paper/2006.10226/) (2020)
- [Efficient Residue Number System Based Winograd Convolution](https://mlsanity.com/paper/2007.12216/) (2020)
- [Pre-Quantized Deep Learning Models Codified in ONNX to Enable Hardware/Software Co-Design](https://mlsanity.com/paper/2110.01730/) (2021)

## Built on by (most cited first, 10 of 10)

- [Open-source FPGA-ML codesign for the MLPerf Tiny Benchmark](https://mlsanity.com/paper/2206.11791/) (2022)
- [hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware](https://mlsanity.com/paper/2512.01463/) (2025)
- [End-to-end codesign of Hessian-aware quantized neural networks for FPGAs and ASICs](https://mlsanity.com/paper/2304.06745/) (2023)
- [NaviSplit: Dynamic Multi-Branch Split DNNs for Efficient Distributed Autonomous Navigation](https://mlsanity.com/paper/2406.13086/) (2024)
- [FINN-GL: Generalized Mixed-Precision Extensions for FPGA-Accelerated LSTMs](https://mlsanity.com/paper/2506.20810/) (2025)
- [Hardware-Aware Feature Extraction Quantisation for Real-Time Visual Odometry on FPGA Platforms](https://mlsanity.com/paper/2507.07903/) (2025)
- [SIRA: Scaled-Integer Range Analysis for Optimizing FPGA Dataflow Neural Network Accelerators](https://mlsanity.com/paper/2508.21493/) (2025)
- [ALADIN: Accuracy-Latency-Aware Design-space Inference Analysis for Embedded AI Accelerators](https://mlsanity.com/paper/2603.08722/) (2026)
- [CADENCE: Context-Adaptive Depth Estimation for Navigation and Computational Efficiency](https://mlsanity.com/paper/2604.07286/) (2026)
- [QuantSR+: Pushing the Limit of Quantized Image Super-Resolution Networks](https://mlsanity.com/paper/2605.22351/) (2026)

## Abstract (arXiv)

We present extensions to the Open Neural Network Exchange (ONNX) intermediate representation format to represent arbitrary-precision quantized neural networks. We first introduce support for low precision quantization in existing ONNX-based quantization formats by leveraging integer clipping, resulting in two new backward-compatible variants: the quantized operator format with clipping and quantize-clip-dequantize (QCDQ) format. We then introduce a novel higher-level ONNX format called quantized ONNX (QONNX) that introduces three new operators -- Quant, BipolarQuant, and Trunc -- in order to represent uniform quantization. By keeping the QONNX IR high-level and flexible, we enable targeting a wider variety of platforms. We also present utilities for working with QONNX, as well as examples of its usage in the FINN and hls4ml toolchains. Finally, we introduce the QONNX model zoo to share low-precision quantized neural networks.