Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures

[2608.14586] Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Distributed, Parallel, and Cluster Computing arXiv:2608.14586 (cs) [Submitted on 19 Jun 2026] Title: Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures Authors: Haibo HU , Lianming Huang , Qiao Li , Nan Guan , Chun Jason Xue View a PDF of the paper titled Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures, by Haibo HU and 4 other authors View PDF HTML (experimental) Abstract: Vision-Language-Action (VLA) models are becoming a promising paradigm for autonomous driving, but their deployment on existing vehicle platforms remains difficult because they introduce both high inference latency and strong GPU-side resource pressure. In a full autonomous driving stack, this problem is even more pronounced: legacy vehicle platforms were provisioned for modular pipelines, yet after several planning-related functions are absorbed into a unified VLA model, part of the original CPU budget becomes underutilized, while the visual encoder and the main reasoning path still concentrate most computation and memory demand on the GPU. As a result, directly deploying VLA together with the rest of the onboard system can be hard under realistic GPU memory constraints. To address this issue, we present a hybrid CPU--GPU inference framework with flexible resource scheduling for autonomous driving. Our design partitions the VLA backbone at the block-layer granularity, executes the visual encoder and LLM prefix on the GPU, and offloads the LLM suffix to the CPU through a cross-frame asynchronous pipeline, thereby exposing a schedulable boundary for redistributing compute and memory pressure across heterogeneous processors. We evaluate the proposed framework on two representative driving VLA models, Orion and MindDrive. On Bench2Drive, our method reduces average latency from 521ms to 408.0ms for Orion and from 443ms to 306.2ms for MindDrive, corresponding to 21.7% and 30.9% reduction, respectively. For Orion, the estimated peak GPU memory is further reduced from 45GB to 29GB. In real-vehicle deployment under coexistence with this http URL , native Orion cannot run because the onboard GPU memory budget is insufficient, whereas the hybrid version runs successfully together with the full vehicle stack. Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2608.14586 [cs.DC] (or arXiv:2608.14586v1 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2608.14586 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Haibo Hu [ view email ] [v1] Fri, 19 Jun 2026 02:37:35 UTC (3,378 KB) Full-text links: Access Paper: View a PDF of the paper titled Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures, by Haibo HU and 4 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.DC < prev | next > new | recent | 2026-08 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies#927297
• Efficient Block-Layer Parallel Inference for Vision-Language-Action on Hybrid Architectures#609026
• VLA Foundry: A Unified Framework for Training Vision-Language-Action Models#124125
• EcoVLA: Energy-Efficient Device-Edge Co-Inference for Vision-Language-Action Models under Real-Time Constraints#616664
• EcoVLA: Energy-Efficient Device-Edge Co-Inference for Vision-Language-Action Models under Real-Time Constraints#616164