RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zang, Hongzhi, Wei, Mingjie, Xu, Si, Wu, Yongji, Guo, Zhen, Wang, Yuanqing, Lin, Hao, Wang, Peihong, Shi, Liangzhi, Xie, Yuqing, Xu, Zhexuan, Liu, Zhihao, Chen, Kang, Tang, Wenhao, Zhang, Quanlu, Zhang, Weinan, Yu, Chao, Wang, Yu
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917256231911424
author Zang, Hongzhi
Wei, Mingjie
Xu, Si
Wu, Yongji
Guo, Zhen
Wang, Yuanqing
Lin, Hao
Wang, Peihong
Shi, Liangzhi
Xie, Yuqing
Xu, Zhexuan
Liu, Zhihao
Chen, Kang
Tang, Wenhao
Zhang, Quanlu
Zhang, Weinan
Yu, Chao
Wang, Yu
author_facet Zang, Hongzhi
Wei, Mingjie
Xu, Si
Wu, Yongji
Guo, Zhen
Wang, Yuanqing
Lin, Hao
Wang, Peihong
Shi, Liangzhi
Xie, Yuqing
Xu, Zhexuan
Liu, Zhihao
Chen, Kang
Tang, Wenhao
Zhang, Quanlu
Zhang, Weinan
Yu, Chao
Wang, Yu
contents Recent advances in vision-language-action (VLA) models have motivated the extension of their capabilities to embodied settings, where reinforcement learning (RL) offers a principled way to optimize task success through interaction. However, existing methods remain fragmented, lacking both a unified platform for fair comparison across architectures and algorithms and an efficient system design for scalable training. To address these challenges, we introduce RLinf-VLA, a unified and efficient framework for scalable RL training of VLA models. RLinf-VLA achieves unification by providing a unified interface that standardizes the integration of diverse VLA architectures, multiple RL algorithms, and heterogeneous simulators, enabling extensibility. To ensure efficiency, the system adopts a flexible resource allocation architecture for rendering, inference, and training workloads in RL pipelines. In particular, for GPU-parallelized simulators, RLinf-VLA introduces a hybrid fine-grained pipeline allocation strategy, yielding a 1.61x-1.88x training speedup. Using this unified system, models trained with RLinf-VLA demonstrate consistent performance improvements of approximately 20-85% across multiple simulation benchmarks, including LIBERO, ManiSkill, and RoboTwin. Furthermore, we distill a set of training practices for effective RL-based VLA training. We position RLinf-VLA as a foundational system to enable efficient, unified, and reproducible research in embodied intelligence.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06710
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models
Zang, Hongzhi
Wei, Mingjie
Xu, Si
Wu, Yongji
Guo, Zhen
Wang, Yuanqing
Lin, Hao
Wang, Peihong
Shi, Liangzhi
Xie, Yuqing
Xu, Zhexuan
Liu, Zhihao
Chen, Kang
Tang, Wenhao
Zhang, Quanlu
Zhang, Weinan
Yu, Chao
Wang, Yu
Robotics
Recent advances in vision-language-action (VLA) models have motivated the extension of their capabilities to embodied settings, where reinforcement learning (RL) offers a principled way to optimize task success through interaction. However, existing methods remain fragmented, lacking both a unified platform for fair comparison across architectures and algorithms and an efficient system design for scalable training. To address these challenges, we introduce RLinf-VLA, a unified and efficient framework for scalable RL training of VLA models. RLinf-VLA achieves unification by providing a unified interface that standardizes the integration of diverse VLA architectures, multiple RL algorithms, and heterogeneous simulators, enabling extensibility. To ensure efficiency, the system adopts a flexible resource allocation architecture for rendering, inference, and training workloads in RL pipelines. In particular, for GPU-parallelized simulators, RLinf-VLA introduces a hybrid fine-grained pipeline allocation strategy, yielding a 1.61x-1.88x training speedup. Using this unified system, models trained with RLinf-VLA demonstrate consistent performance improvements of approximately 20-85% across multiple simulation benchmarks, including LIBERO, ManiSkill, and RoboTwin. Furthermore, we distill a set of training practices for effective RL-based VLA training. We position RLinf-VLA as a foundational system to enable efficient, unified, and reproducible research in embodied intelligence.
title RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models
topic Robotics
url https://arxiv.org/abs/2510.06710