aiqbench analysis
This paper presents VQVLA, an algorithm-hardware co-design framework that accelerates Vision-Language-Action (VLA) models by dynamically adjusting quantization precision based on robot execution states. By reusing centroids and exploiting weight similarity, the approach achieves significant speedups over standard GPUs while maintaining task success rates with negligible accuracy loss. The work addresses critical inference latency bottlenecks in embodied AI systems through specialized hardware support for dynamic precision selection.
A Motion-Aware Vector Quantization Framework with Centroid Reuse for Efficient VLA Inference
arXiv:2607.24148v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have demonstrated strong potential for embodied AI, yet their high inference latency on GPUs limits real-time deployment. Existing accelerators, such as Dadu-Corki, improve efficiency but treat VLA models as full-precision workloads, leaving substantial redundancy in both memory and computation underexploited. In this paper, we propose VQVLA, an algorithm-hardware co-design framework that accelerates VLA inference by exploiting weight similarity and execution dynamics. We first introduce MotionVQ, a motion-aware vector quantization scheme that dynamically adjusts quantization precision based on the robot's execution state, reducing memory access while preserving task success rate. We then propose a merged-centroid vectorized GEMM paradigm that operates on the codebook-index representation, eliminating redundant multiplications through spatial aggregation and temporal reuse of centroids. To realize these optimizations, we design an accelerator that efficiently supports dynamic precision selection and centroid-reuse computation. Experimental results show that VQVLA achieves 6.5x, 2.8x, 1.9x, 3.3x, and 4.3x speedup over the A100 GPU, Dadu-Corki, LUT-DLA, CodeGEMM, and ShiftAddLLM, respectively, with negligible accuracy degradation.
Abstract:Vision-Language-Action (VLA) models have demonstrated strong potential for embodied AI, yet their high inference latency on GPUs limits real-time deployment. Existing accelerators, such as Dadu-Corki, improve efficiency but treat VLA models as full-precision workloads, leaving substantial redundancy in both memory and computation underexploited. In this paper, we propose VQVLA, an algorithm-hardware co-design framework that accelerates VLA inference by exploiting weight similarity and execution dynamics. We first introduce MotionVQ, a motion-aware vector quantization scheme that dynamically adjusts quantization precision based on the robot's execution state, reducing memory access while preserving task success rate. We then propose a merged-centroid vectorized GEMM paradigm that operates on the codebook-index representation, eliminating redundant multiplications through spatial aggregation and temporal reuse of centroids. To realize these optimizations, we design an accelerator that efficiently supports dynamic precision selection and centroid-reuse computation. Experimental results show that VQVLA achieves 6.5x, 2.8x, 1.9x, 3.3x, and 4.3x speedup over the A100 GPU, Dadu-Corki, LUT-DLA, CodeGEMM, and ShiftAddLLM, respectively, with negligible accuracy degradation.
Artificial Intelligence (cs.AI)
arXiv:2607.24148 [cs.AI]
(or arXiv:2607.24148v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2607.24148 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Full text under the publisher's open licence · 255 words
Related coverage
Productionizing TurboQuant on AMD GPUs for KV-Cache-Bound LLM Inference
AMD · Jun 11, 2026
PagedAttention vs. RadixAttention: Optimizing LLM KV Cache Management
Analytics Vidhya · Aug 21, 2026
The hard physics and complex economics of AI’s insatiable hunger for power
SiliconANGLE · Sep 11, 2026
Unlocking Sim2Real for a Robotic Arm with RL Accelerated by AMD Instinct GPUs
AMD · Sep 29, 2026