Metraipolis › archive › dc_1fbebcf9cc35

Beyond LLM Serving: Characterizing Vision-Language-Action Workloads for Embodied AI System Design — arXiv:2610.05062

Source
arXiv
Original
https://arxiv.org/abs/2610.05062
License (as archived)
CC0-1.0
Archived
2026-10-06T05:29:09.822Z
SHA-256 of the archived text
fe39edbfc8ae2ea4f9aff12aea3f7414d7ced9c758f30109f03b50ba0d162c67
Stable URL of this page
https://metraipolis.com/documento/dc_1fbebcf9cc35

How to cite

Beyond LLM Serving: Characterizing Vision-Language-Action Workloads for Embodied AI System Design — arXiv:2610.05062 — arXiv, https://arxiv.org/abs/2610.05062 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1fbebcf9cc35 (Metraipolis archive dc_1fbebcf9cc35, sha256 fe39edbfc8ae2ea4f9aff12aea3f7414d7ced9c758f30109f03b50ba0d162c67)
BibTeX
@misc{dc1fbebcf9cc35,
  title = {Beyond LLM Serving: Characterizing Vision-Language-Action Workloads for Embodied AI System Design — arXiv:2610.05062},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05062},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1fbebcf9cc35; sha256 fe39edbfc8ae2ea4f9aff12aea3f7414d7ced9c758f30109f03b50ba0d162c67},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05062. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models translate multimodal observations into low-level robot actions. During robot operation, each control period sets an inference deadline, and overruns leave the robot acting on stale observations, reducing task success. Meeting this deadline motivates on-device or nearby edge execution, where a single robot requires batch-1 inference outside the design point of LLM serving systems. Although VLA architectures combine familiar vision-language, autoregressive, and diffusion-style components, their runtime behavior in this batch-1 control setting remains uncharacterized. We characterize four representative VLA models on an edge GPU server and two onboard SoCs, using single-inference profiling and 43,200 closed-loop episodes. Action tensor dimensionality determines whether a stage is memory- or compute-bound, platform balance can shift that bottleneck, and GPU frequency scaling yields a platform-dependent energy-latency sweet spot. In closed-loop operation, overlapping inference with action execution creates an accuracy-speed-energy tradeoff, and no configuration is Pareto-dominant across deployment SLOs. These results guide joint design of VLA model architectures, hardware, and runtime policies.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.AR, nuovi lavori del giorno).
Categorie arXiv: Hardware Architecture (cs.AR) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.05062 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:24:50.290Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).