OSCAR: An Open-Source Flexible and Hierarchical AI Accelerator Generator with Accurate Power Model
Open Access
- 27 August 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
Abstract
With the growing demand for artificial intelligence (AI) applications, high-performance and energy-efficient AI chips are needed to support the computation. However, architectural-level AI chip design, PPA evaluation, and power estimation remain challenging due to the exponential set of possible designs and the difficulty of accurately modeling the impact of diverse dataflows and workflows on the underlying hardware at the architectural-design stage. We propose a novel open-source framework named OSCAR, which, given a set of hardware and workload specifications, provides architecture-level power estimation and can also automatically generate Chisel and synthesizable RTL of the custom AI chip. Our contributions include (1) a flexible and hierarchical AI chip design space, software and hardware stack, and an RTL generator supporting dense, Transformer, Winograd, systolic, and reconfigurable architectures in one unified framework, (2) hierarchy-based data-sensitive power model using architectural-level toggling features, achieving 3.8% error and correlation coefficient R > 0.99 to post-synthesis power, outperforming state-of-the-art power estimation methods, (3) validation of our power model by performing design space exploration, finding designs with better Pareto-optimality, 2.5 × lower power or 2 × better runtime metrics, compared with using prior art power models, (4) a tape-out of an AI chip based on DSE results, with OSCAR modeling its power with over 90% accuracy.Keywords
This publication has 21 references indexed in Scilit:
- MBS: A High-Precision Approximation Method for Softmax and Efficient Hardware ImplementationIEEE Transactions on Circuits and Systems I: Regular Papers, 2025
- A Tiny Accelerator for Mixed-Bit Sparse CNN Based on Efficient Fetch Method of SIMO SPadIEEE Transactions on Circuits and Systems II: Express Briefs, 2023
- Sparseloop: An Analytical Approach To Sparse Tensor Accelerator Modeling2022
- Gemmini: Enabling Systematic Deep-Learning Architecture Evaluation via Full-Stack Integration2021
- Integrating NVIDIA Deep Learning Accelerator (NVDLA) with RISC-V SoC on FireSim2019
- Cambricon-S: Addressing Irregularity in Sparse Neural Networks through A Cooperative Software/Hardware Approach2018
- Deep Residual Learning for Image Recognition2016
- Computer Arithmetic2012
- Networks on chips: a new SoC paradigmComputer, 2002
- CACTI: an enhanced cache access and cycle time modelIEEE Journal of Solid-State Circuits, 1996