MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Driving

TL;DR

MindVLA-U1 introduces a unified streaming VLA architecture for autonomous driving that outperforms human drivers and state-of-the-art models by integrating semantic reasoning, temporal context, and continuous control in a single, efficient system.

Contribution

This paper presents the first unified streaming VLA architecture that processes driving video and language commands in a single pass, improving planning accuracy and latency.

Findings

Surpasses experienced human drivers on WOD-E2E benchmark with 8.20 RFS.

Achieves state-of-the-art planning ADEs over prior VA/VLA models.

Matches VA latency at 16 FPS while maintaining natural language interfaces.

Abstract

Autonomous driving has progressed from modular pipelines toward end-to-end unification, and Vision-Language-Action (VLA) models are a natural extension of this journey beyond Vision-to-Action (VA). In practice, driving VLAs have often trailed VA on planning quality, suggesting that the difficulty is not simply model scale but the interface through which semantic reasoning, temporal context, and continuous control are combined. We argue that this gap reflects how VLA has been built -- as isolated subtask improvements that fail to compose coherent driving capabilities -- rather than what VLA is. We present MindVLA-U1, the first unified streaming VLA architecture for autonomous driving. A unified VLM backbone produces AR language tokens (optional) and flow-matching continuous action trajectories in a single forward pass over one shared representation, preserving the natural output form of each modality. A full streaming design processes the driving video framewise rather than as fixed video-action chunks under costly temporal VLM modeling. Planned trajectories evolve smoothly across frames while a learned streaming memory channel carries temporal context and updates. The unified architecture enables fast/slow systems on dense & sparse MoT backbones via flexible self-attention context management, and exposes a measurable language-control path for action: language-predicted driving intents steers the action diffusion via classifier-free guidance (CFG), turning language-side intent into control signals for continuous action planning. On the long-tail WOD-E2E benchmark, MindVLA-U1 surpasses experienced human drivers for the first time (8.20 RFS vs. 8.13 GT RFS) with 2 diffusion steps, achieves state-of-the-art planning ADEs over prior VA/VLA by large margins, and matches VA latency (16 FPS vs. RAP's 18 FPS at 1B scale) while preserving natural language interfaces for human-vehicle interaction.