OpenVLA: An Open-Source Vision-Language-Action Model for Robot Control

ArticleMoo Jin Kim et al.Research

Researchers from Stanford, UC Berkeley, Google DeepMind, and partner institutions introduce OpenVLA, an open-source 7-billion-parameter vision-language-action model. Trained on 970k real-world robot demonstrations, it surpasses closed models such as RT-2-X while enabling parameter-efficient fine-tuning on consumer-grade hardware.

Key points

  1. OpenVLA integrates a Llama 2 language backbone with visual feature representations fused from DINOv2 and SigLIP.
  2. Pretrained on 970k manipulation trajectories from the Open X-Embodiment dataset, supporting multiple robot embodiments out of the box.
  3. Outperforms the 55B-parameter RT-2-X model by 16.5% in absolute task success rate across 29 evaluation tasks, despite having seven times fewer parameters.
  4. Surpasses from-scratch imitation learning baselines like Diffusion Policy by 20.4% in multi-task, multi-object environments requiring language grounding.
  5. Supports efficient adaptation via Low-Rank Adaptation (LoRA) and post-training quantization, allowing execution and fine-tuning on consumer GPUs without performance drops.
  6. Model checkpoints, training code, and adaptation notebooks are fully released on Hugging Face.

Why it matters

Most prior vision-language-action models were proprietary and inaccessible, with unclear adaptation paths for custom setups. OpenVLA bridges this gap by providing an open, state-of-the-art policy that makes advanced robotic fine-tuning accessible on standard commodity hardware.

Analysis

OpenVLA highlights the efficacy of formulating visuomotor control directly within standard multimodal LLM architectures by mapping robot actions to vocabulary tokens. By blending spatial representations from DINOv2 with semantic cues from SigLIP, the model generalises across disparate embodiments. Its open-source release, coupled with validated LoRA workflows on consumer GPUs, provides a practical foundation for reproducible robotics research.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Link:Moo Jin Kim, Karl Pertsch, Siddharth Karamcheti et al.

    OpenVLA: Open-Source Vision-Language-Action Model for Robotics

    OpenVLA is an open-source 7B parameter vision-language-action (VLA) model designed for generalist robot manipulation. Built on Prismatic-7B and Llama 2, it was trained on 970k episodes from the Open X-Embodiment dataset and outperforms earlier generalist policies such as Octo, RT-1-X, and in several benchmarks the closed 55B RT-2-X.

    AI & LLMs· Research

  • Link:Zhe Liu, Jinghua Hou, Yuxiang Lu, et al.

    StreamPI: Streaming Multimodal Temporal Modeling for Vision-Language-Action Models

    Researchers from The University of Hong Kong and ACE Robotics introduce StreamPI, a framework that equips single-frame Vision-Language-Action (VLA) models such as π0.5 with temporal reasoning without adding new parameters. By bundling images and instructions into atomic temporal units and using random-interval training, StreamPI improves precision, memory, and robustness in real-robot manipulation.

    AI & LLMs· Research

  • Article:Peiyan Li, Yuze Zhu, Yixiang Chen, Qisen Ma, Yuan Xu, Jiabing Yang, He Guan, Yan Huang, Hongtao Wu, Xiao Ma, Tao Kong, Liang Wang, Tieniu Tan

    BridgeVLA++: A Memory-Augmented 3D Vision-Language-Action Framework for Robotics

    Researchers introduce BridgeVLA++, an upgraded vision-language-action framework designed for 3D robotic manipulation. It introduces a unified spatio-temporal memory architecture into pre-trained vision-language models, enabling persistent spatial reasoning and temporal history tracking without losing data efficiency.

    AI & LLMs· Research

  • Link:Dyna Robotics

    Dyna Robotics Unveils DYNA-2: World-Action Model Driven by Human Video Data

    Dyna Robotics has launched DYNA-2, a World-Action Model (WAM) pre-trained on more than one million hours of egocentric human video. The model circumvents traditional teleoperation bottlenecks, transferring physical reasoning and action policies directly to diverse robotic hardware.

    AI & LLMs· Announcement

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.