Argus: Quality-Aware High-Throughput Text-to-Image Inference Serving System

TL;DR

Argus is a system that improves text-to-image inference throughput by dynamically selecting approximation levels for each prompt, reducing latency violations and increasing quality and throughput.

Contribution

It introduces a novel approach to adaptively choose approximation strategies for T2I models, balancing quality and throughput in high-demand settings.

Findings

Achieves 10x fewer latency SLO violations

Provides 10% higher average quality

Increases throughput by 40%

Abstract

Text-to-image (T2I) models have gained significant popularity. Most of these are diffusion models with unique computational characteristics, distinct from both traditional small-scale ML models and large language models. They are highly compute-bound and use an iterative denoising process to generate images, leading to very high inference time. This creates significant challenges in designing a high-throughput system. We discovered that a large fraction of prompts can be served using faster, approximated models. However, the approximation setting must be carefully calibrated for each prompt to avoid quality degradation. Designing a high-throughput system that assigns each prompt to the appropriate model and compatible approximation setting remains a challenging problem. We present Argus, a high-throughput T2I inference system that selects the right level of approximation for each prompt to maintain quality while meeting throughput targets on a fixed-size cluster. Argus intelligently switches between different approximation strategies to satisfy both throughput and quality requirements. Overall, Argus achieves 10x fewer latency service-level objective (SLO) violations, 10% higher average quality, and 40% higher throughput compared to baselines on two real-world workload traces.