INAR-VL: Input-Aware Routing for Edge-Cloud Vision-Language Inference

TL;DR

INAR-VL is a lightweight routing system that intelligently directs vision-language inference tasks between edge and cloud to optimize latency, energy, and accuracy in heterogeneous environments.

Contribution

It introduces a novel input-aware routing approach that dynamically balances edge and cloud execution for multimodal inference, improving efficiency and accuracy.

Findings

Executes 36% of requests on edge

Reduces latency by 24%

Lowers energy consumption by 26%

Abstract

Edge deployment of Vision-Language Models (VLMs) faces a tradeoff between latency and accuracy: cloud execution provides high-quality predictions but incurs communication delay and energy cost, while edge-only execution is faster but less accurate due to limited model capacity. This trade-off is further complicated by heterogeneity in image quality and reasoning complexity, making static placement suboptimal. We present INAR-VL, a lightweight edge-cloud routing system for multimodal inference in a two-tier deployment. INAR-VL maintains complementary VLMs across edge and cloud and uses lightweight image and text complexity signals to guide routing and model selection, executing simple queries locally while offloading complex ones when beneficial. Evaluation on visual question answering shows that INAR-VL executes 36% of requests on the edge, reduces latency by 24%, lowers energy by 26%, and preserves 97% of cloud-level accuracy.