Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference
Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood. Existing efficiency efforts focus predominantly on reducing visual tokens, implicitly treating visual processing as the dominant energy cost. We overturn this implicit assumption through the first systematic energy profiling of on-device VLM inference, spanning five models across three architecture families, four input resolutions, and two hardware platforms (NVIDIA RTX 3070 and Jetson Orin NX). Our analysis yields three findings. First, average infer
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 58%How NVIDIA’s Inference Software Stack Powers the Lowest Token Cost →
- PossiblePossibly related (embedding) · 53%NVIDIA Unlocks AI Compute at Scale, Inviting Partners to Power the AI Infrastructure Buildout →
- PossiblePossibly related (embedding) · 53%How Outpost VFX Uses AWS to Accelerate AI Model Training for Visual Effects →
- PossiblePossibly related (embedding) · 52%open-edge-platform/geti →
- PossiblePossibly related (embedding) · 52%NVIDIA Unlocks AI Compute at Scale, Inviting Capital Partners to Power the AI Infrastructure Buildout →
- LinkedLinked via arxiv author · 85%Junfei Zhan →
“Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference”
- LinkedLinked via arxiv author · 85%Haoxun Shen →
“Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference”
- LinkedLinked via arxiv author · 85%Mingang Guo →
“Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference”