FluxVLA Engine: A One-Stop VLA Engineering Platform for Embodied Intelligence
cs.RO, cs.AI
Submitted: 2026-09-15
Updated: 2026-09-15
Code: https://github.com/FluxVLA/FluxVLA
License: http://creativecommons.org/licenses/by/4.0/
The gist: Vision-language-action (VLA) models, world-action models (WAMs), and offline reinforcement learning methods are rapidly expanding the design space of embodied policies, yet turning these algorithms
Terminology
Abstract
Vision-language-action (VLA) models, world-action models (WAMs), and offline reinforcement learning methods are rapidly expanding the design space of embodied policies, yet turning these algorithms into reliable robot systems remains constrained by fragmented data formats, training stacks, evaluation protocols, inference runtimes, and embodiment-specific interfaces. We present FluxVLA Engine, an open, configuration-driven platform that turns heterogeneous embodied-policy components into a reproducible data-to-deployment workflow. Rather than introducing another policy model, FluxVLA standardizes interfaces for datasets, visual-language and world models, action heads, reward- or advantage-weighted learning, distributed training, simulation evaluation, optimized inference, and robot operators. The engine further integrates compositional dual-arm simulation, scalable automatic data generation, and model-decoupled human-in-the-loop rollout, takeover, correction collection, and reward annotation. For responsive physical execution, it combines Real-Time Chunking (RTC) with accelerated inference backends, lightweight remote GPU serving, and configurable trajectory post-processing. Together, these capabilities connect offline learning, simulation validation, online correction, and real-robot execution through shared and auditable contracts. FluxVLA therefore targets the engineering bottlenecks separating promising embodied-learning algorithms from reproducible evaluation and dependable deployment. Code is available at https://github.com/FluxVLA/FluxVLA
Sources
- OpenVLA: An Open-Source Vision-Language-Action Model
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
- World Action Models are Zero-shot Policies
- FAST: Efficient Action Tokenization for Vision-Language-Action Models
- DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control
- Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
- Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
- Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
- MMDetection: Open MMLab Detection Toolbox and Benchmark
- Octo: An Open-Source Generalist Robot Policy
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
- SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
- Fast-WAM: Do World Action Models Need Test-time Future Imagination?
- RT-1: Robotics Transformer for Real-World Control at Scale
- ARM: Advantage Reward Modeling for Long-Horizon Manipulation
- RLDS: an Ecosystem to Generate, Share and Use Datasets in Reinforcement Learning
- Open X-Embodiment: Robotic Learning Datasets and RT-X Models
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving