# InstMeter: An Instruction-Level Method to Predict Energy and Latency of DL Model Inference on MCUs

arXiv: https://arxiv.org/abs/2603.04134 · HTML: https://mlsanity.com/paper/2603.04134/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 45 references that are in the map.
- First posted on arXiv 4 Mar 2026.

## Builds on (most cited first, 13 of 13)

- [ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware](https://mlsanity.com/paper/1812.00332/) (2018)
- [NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search](https://mlsanity.com/paper/2001.00326/) (2020)
- [TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systems](https://mlsanity.com/paper/2010.08678/) (2020)
- [BRP-NAS: Prediction-based NAS using GCNs](https://mlsanity.com/paper/2007.08668/) (2020)
- [A Flexible Framework for Multi-Objective Bayesian Optimization using Random Scalarizations](https://mlsanity.com/paper/1805.12168/) (2018)
- [μNAS: Constrained Neural Architecture Search for Microcontrollers](https://mlsanity.com/paper/2010.14246/) (2020)
- [Cross-Language Binary-Source Code Matching with Intermediate Representations](https://mlsanity.com/paper/2201.07420/) (2022)
- [Towards Battery-Free Machine Learning and Inference in Underwater Environments](https://mlsanity.com/paper/2202.08174/) (2022)
- [BinaryAI: Binary Software Composition Analysis via Intelligent Binary Source Code Matching](https://mlsanity.com/paper/2401.11161/) (2024)
- [BinPro: A Tool for Binary Source Code Provenance](https://mlsanity.com/paper/1711.00830/) (2017)
- [MONAS: Efficient Zero-Shot Neural Architecture Search for MCUs](https://mlsanity.com/paper/2408.15034/) (2024)
- [MAPLE-Edge: A Runtime Latency Predictor for Edge Devices](https://mlsanity.com/paper/2204.12950/) (2022)
- [Accurate Energy Modelling on the Cortex-M0 Processor for Profiling and Static Analysis](https://mlsanity.com/paper/2301.12806/) (2023)

## Built on by (most cited first, 2 of 2)

- [CREST: Deployment-Realistic Hardware-in-the-Loop NAS for Embedded Sensing Systems](https://mlsanity.com/paper/2606.15004/) (2026)
- [EdgeCraft: Automated Model Crafting for Edge IoT](https://mlsanity.com/paper/2609.35167/) (2026)

## Abstract (arXiv)

Deep learning (DL) models can now run on microcontrollers (MCUs). Through neural architecture search (NAS), we can search DL models that meet the constraints of MCUs. Among various constraints, energy and latency costs of the model inference are critical metrics. To predict them, existing research relies on coarse proxies such as multiply-accumulations (MACs) and model's input parameters, often resulting in inaccurate predictions or requiring extensive data collection. In this paper, we propose InstMeter, a predictor leveraging MCUs' clock cycles to accurately estimate the energy and latency of DL models. Clock cycles are fundamental metrics reflecting MCU operations, directly determining energy and latency costs. Furthermore, a unique property of our predictor is its strong linearity, allowing it to be simple and accurate. We thoroughly evaluate InstMeter under different scenarios, MCUs, and software settings. Compared with state-of-the-art studies, InstMeter can reduce the energy and latency prediction errors by 3× and 6.5×, respectively, while requiring 100× and 10× less training data. In the NAS scenario, InstMeter can fully exploit the energy budget, identifying optimal DL models with higher inference accuracy. We also evaluate InstMeter's generalization performance through various experiments on three ARM MCUs (Cortex-M4, M7, M33) and one RISC-V-based MCU (ESP32-C3), different compilation options (-Os, -O2), GCC versions (v7.3, v10.3), application scenarios (keyword spotting, image recognition), dynamic voltage and frequency scaling, temperatures (21degC, 43degC), and software settings (TFLMv2.4, TFLMvCI). We will open our source codes and the MCU-specific benchmark datasets.