Implementation and Evaluation of Stable Diffusion on a General-Purpose CGLA Accelerator

TL;DR

This paper implements and evaluates stable diffusion image generation on a general-purpose CGRA accelerator, demonstrating promising performance and power efficiency, and providing guidelines for future AI accelerator designs.

Contribution

It presents the first implementation and detailed evaluation of stable diffusion kernels on a versatile CGRA platform, guiding future AI hardware development.

Findings

IMAX3 achieves promising performance on FPGA prototype

Power efficiency is high, especially in projected ASIC form

Provides design guidelines for next-generation AI accelerators

Abstract

This paper presents the first implementation and in-depth evaluation of the primary computational kernels from the stable-diffusion.cpp image generation framework on IMAX3, a general-purpose Coarse-Grained Reconfigurable Array (CGRA) accelerator. We designed IMAX3 as a versatile computational platform, and this work assesses its capabilities by executing a demanding image generation workload. We evaluate its performance on a current Field-Programmable Gate Array (FPGA) prototype to establish a baseline and project its potential for a future Application-Specific Integrated Circuit (ASIC) implementation. Our results demonstrate that, despite its general-purpose architecture, IMAX3 achieves promising performance and power efficiency, particularly in its projected ASIC form. This work provides concrete guidelines for future IMAX architectural designs and establishes a foundation for developing next-generation, AI-specialized Coarse-Grained Linear Array (CGLA) accelerators by refining this versatile platform. Ultimately, this achievement contributes to the realization of energy-efficient, on-device, multi-modal AI platforms.