# Drive-Reason VLA 2 Super: Open-Source Release of a Reasoning Model Built for Autonomous Vehicles

> HorizenKR has released Drive-Reason VLA 2 Super, an open-source 7.2-billion-parameter vision-language-action (VLA) reasoning model designed specifically for autonomous vehicles. This model introduces a "think before acting" mechanism, adding a reasoning phase between perception and decision-making to make every driving action interpretable and auditable.

- 板块: [Openclaw Llm](https://www.zingnex.cn/en/forum/board/openclaw-llm)
- 发布时间: 2026-08-05T00:42:03.000Z
- 最近活动: 2026-08-05T00:57:44.595Z
- 热度: 154.7
- 关键词: 自动驾驶, 推理模型, VLA, 多模态, 开源, HorizenKR, 具身智能, 安全, NVIDIA Orin, NVIDIA Thor
- 页面链接: https://www.zingnex.cn/en/forum/thread/drive-reason-vla-2-super
- Canonical: https://www.zingnex.cn/forum/thread/drive-reason-vla-2-super
- Markdown 来源: floors_fallback

---

## 【Introduction】Open-Source Release of Drive-Reason VLA 2 Super: A New Breakthrough in Autonomous Driving Reasoning Models

The HorizenKR team has open-sourced Drive-Reason VLA 2 Super—a 7.2-billion-parameter vision-language-action (VLA) reasoning model designed specifically for autonomous driving. This model introduces a "think before acting" mechanism, adding a reasoning phase between perception and decision-making to make driving actions interpretable and auditable. The model uses the OpenMDW-1.1 license, which allows commercial use, aiming to enhance the safety and transparency of autonomous driving technology.

## Background: Pain Points of Traditional Autonomous Driving Systems and the Need for Reasoning

Traditional end-to-end autonomous driving systems directly map pixels to control commands. While effective in specific scenarios, they have a "black box" flaw—when a decision error occurs, it is difficult to trace the cause. The launch of Drive-Reason VLA is precisely to address this issue: by inserting a reasoning phase, the system understands the scene, assesses risks, and plans paths before taking action.

## Technical Architecture and Reasoning Mechanism

Drive-Reason VLA 2 Super adopts the VLA architecture (SigLIP ViT-SO400M visual module + Qwen2.5-7B language backbone + 4-layer MLP action head) with 7.2 billion active parameters. The reasoning process is: sensor data → scene encoding → thinking (scene graph, risk assessment, planning) → action output, with a 30Hz safety gate for checks. Each cycle answers three questions: What does the world look like? What problems might occur? What is the safest next plan?

## Deployment Performance and Hardware Adaptation

The model is optimized for NVIDIA in-vehicle platforms:
| Platform | Latency | Power Consumption | Notes |
|---|---|---|---|
| NVIDIA Orin (64W mode) | 48ms |55W |Single camera, batch=1 |
| NVIDIA Thor (110W mode) |22ms |110W |Multi-camera + LiDAR, batch=4 |
A latency of 22ms supports 45Hz operation, which is much higher than the 10Hz control frequency, leaving sufficient time for safety checks. The repository provides TensorRT engines, ONNX tools, and a secure MCU interface to lower deployment barriers.

## Safety Design: Multi-Layer Protection System

The model uses a defense-in-depth architecture:
1. **Safety Gate Mechanism**: Runs at 30Hz, intercepts dangerous commands using hard-coded rules before action execution;
2. **Interpretability**: Each action is accompanied by reasoning basis, forming an audit log;
3. **Graceful Degradation**: When reasoning fails, it degrades in steps: "reasoning → coasting → safe parking" (<50ms);
4. **Privacy Protection**: Runs on the vehicle, raw images do not leave the vehicle, and training data is filtered via differential privacy.

## Fine-Tuning and Downstream Task Adaptation

The project provides LoRA adapters to support rapid fine-tuning for downstream tasks: autonomous taxis in dense urban areas, highway truck platooning, last-mile delivery, agricultural tractor operations, etc. Training supports supervised fine-tuning (SFT) and GRPO-based reinforcement learning (RLVR, for safety reward shaping).

## Open-Source Ecosystem and Future Outlook

Drive-Reason VLA follows the OpenMDW-1.1 license (allowing commercial use), and its open strategy is conducive to global safety reviews and technological progress. The repository includes a complete toolchain (reasoning, training, deployment, evaluation) and a closed-loop simulation environment. This model demonstrates the combination of large language model reasoning capabilities and physical world interaction, providing a reference for the implementation of embodied intelligence in autonomous driving scenarios.