---
title: DM05-Vla-Arena
canonical_url: "https://www.modelscope.cn/models/Dexmal/DM05-Vla-Arena"
md_url: "https://www.modelscope.cn/models/Dexmal/DM05-Vla-Arena.md"
repository: Dexmal/DM05-Vla-Arena
last_updated: 2026-08-05
license: gemma
model_type:
  - dm05
architectures:
  - DM05ForConditionalGeneration
base_model:
  - Dexmal/DM05
base_model_relation: finetune
parameters: 5.8B
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 13
stars: 0
tags:
  - robotics
  - robot-control
  - vision-language-action
  - vla
  - dm05
  - dm0.5
  - vla-arena
  - opendm
---

# DM05-Vla-Arena

> DM05-Vla-Arena - Dexmal 在 ModelScope 开源的模型。DM05-vla-arena is the VLA-Arena fine-tuned checkpoint of DM0.5, Dexmal's open-world Vision-Language-Action foundation model for embodied intelligence. DM0.5 uses a Gemma3 4B vision-language backbone with a 680M…

Dexmal/DM05-Vla-Arena 是 ModelScope 魔搭社区上的 5.8B 参数机器学习模型，采用 gemma 许可，基于 Dexmal/DM05 构建。

- **Repository**: Dexmal/DM05-Vla-Arena
- **License**: gemma
- **Parameters**: 5.8B
- **Base model**: Dexmal/DM05
- **Tags**: robotics, robot-control, vision-language-action, vla, dm05, dm0.5, vla-arena, opendm
- **Downloads**: 13
- **Stars**: 0
- **Last updated**: 2026-08-05

Source: https://www.modelscope.cn/models/Dexmal/DM05-Vla-Arena

---

# DM05-vla-arena

![DM0.5](https://raw.githubusercontent.com/dexmal/opendm/main/docs/image/header.png)

<p align="center">
  <a href="https://www.dexmal.com/blog/dm0.5/index_en.html"><img src="https://img.shields.io/badge/📖-Tech_Blog-blue" alt="Tech Blog"></a>
  <a href="https://github.com/dexmal/opendm"><img src="https://img.shields.io/badge/GitHub-OpenDM-181717?logo=github" alt="GitHub"></a>
  <a href="https://huggingface.co/datasets/Dexmal/vla_arena_L0_L"><img src="https://img.shields.io/badge/Dataset-VLA--Arena-0EA5E9?logo=huggingface" alt="VLA-Arena Dataset"></a>
  <a href="https://github.com/PKU-Alignment/VLA-Arena"><img src="https://img.shields.io/badge/GitHub-VLA--Arena-181717?logo=github" alt="VLA-Arena"></a>
  <a href="https://maas.dexmal.com/"><img src="https://img.shields.io/badge/MaaS-Online-brightgreen.svg" alt="MaaS"></a>
</p>

## Introduction

DM05-vla-arena is the VLA-Arena fine-tuned checkpoint of DM0.5, Dexmal's open-world Vision-Language-Action foundation model for embodied intelligence. DM0.5 uses a Gemma3 4B vision-language backbone with a 680M Action Expert to generate continuous robot actions, and is designed for natural-language manipulation, zero-shot generalization, efficient downstream fine-tuning, long-horizon historical context, robust policy behavior, and transfer across robot embodiments.

#### VLA-Arena Results

| Method | Metric | L0 | L1 | L2 |
| --- | --- | ---: | ---: | ---: |
| DM0.5 | SR | 89.0% | 53.6% | 44.1% |
| | CC | 1.9 | 18.75 | 22.52 |

## Quick Start

We recommend using Docker to set up the runtime environment first, which helps avoid version mismatches across CUDA, PyTorch, flash-attn, and other dependencies on the host machine.

### Requirements

```text
System requirements:
Ubuntu 20.04 / 22.04
NVIDIA GPU
NVIDIA Driver
Docker
NVIDIA Container Toolkit
Conda (optional, only required for local pip installation)

Recommended GPUs:
RTX 4090, A100, H100, H20
8 GPUs are recommended for training, and 1 GPU is sufficient for deployment inference.
```

### Docker Installation

```bash
git clone https://github.com/dexmal/opendm.git
cd opendm

docker run -it --rm --gpus all --network host \
  --name opendm \
  --shm-size=16g \
  -v "$PWD":/app/opendm \
  -w /app/opendm \
  dexmal/opendm:latest /bin/bash

# Run from the OpenDM repository root inside the container.
conda activate opendm
pip install -e .
```

### Local Installation

```bash
conda create -n opendm python=3.10 -y
conda activate opendm

pip install torch torchvision \
  --index-url https://download.pytorch.org/whl/cu128

pip install ninja packaging
MAX_JOBS=2 pip install flash-attn --no-build-isolation

# Enter the OpenDM repository root.
cd opendm
pip install -e .
```

## VLA-Arena Testing

Use the VLA-Arena-specific experiment configuration when evaluating this
checkpoint. The inference service and evaluation client run separately and
communicate over HTTP.

For the complete training and evaluation workflow, see the
[DM05 VLA-Arena Training and Evaluation Guide](https://github.com/dexmal/opendm/blob/main/docs/en/dm05_vla_arena.md).

### Start the VLA-Arena Inference Service

Run this command from the OpenDM repository root. The checkpoint must contain
the matching `norm_stats.json`.

```bash
script/dm05_launcher.sh \
  --exp playground/dm05_vla_arena.py \
  --task inference \
  --nproc_per_node 1 \
  --model-config.model-name-or-path ./checkpoints/DM05-vla-arena-bf16 \
  --inference-config.output-action-dim 7
```

Keep the service running while executing the benchmark.

### Prepare the VLA-Arena Evaluation Client

```bash
git clone https://github.com/PKU-Alignment/VLA-Arena
cd VLA-Arena
```

Install the VLA-Arena dependencies:

```bash
sudo apt-get install -y libosmesa6-dev libglfw3 libgl1-mesa-glx libglib2.0-0

pip install robosuite==1.5.1 bddl numpy==1.26.4 requests tqdm pyyaml "imageio[ffmpeg]" pillow

pip install -e .

export MUJOCO_GL=osmesa
```

Copy the evaluation files from the OpenDM repository:

```bash
mkdir -p vla_arena/models/DM05

# Copy evaluation files from the OpenDM repository
cp <path-to-opendm>/third_party/vla_arena/eval.py vla_arena/models/DM05/eval.py
cp <path-to-opendm>/third_party/vla_arena/eval_config.yaml vla_arena/models/DM05/eval_config.yaml
```

Edit `vla_arena/models/DM05/eval_config.yaml` so that `server_url` points to
the running inference service:

```yaml
server_url: "http://<SERVER_IP>:7891/process_frame"
request_timeout: 30

action_horizon: 20
replan_steps: 10
robot_type: "Franka"
batch_size: 1
speed: "0.5"

# "all" evaluates all 170 tasks (11 suites × 3 levels)
task_suite_name: "all"
task_level: 0

num_trials_per_task: 10
seeds: [7, 42, 1000]

model_name: "DM05"
local_log_dir: "./experiments/eval_results"
save_video_mode: "first_success_failure"
```

### Run the VLA-Arena Benchmark

From the VLA-Arena repository root, run:

```bash
python -m vla_arena.models.DM05.eval \
  --config vla_arena/models/DM05/eval_config.yaml \
  --output-dir ./experiments/eval_results
```

Evaluation artifacts are written under `<output-dir>/seed_<N>/`, including
`results_<timestamp>.json`, `tasks_<timestamp>.csv`, and rollout videos under
`videos/`.

## Community and Support

- Learn more about Dexmal products and model updates on the [Dexmal website](https://www.dexmal.com/).
- If you encounter issues, please report them through [GitHub Issues](https://github.com/dexmal/opendm/issues).
- For further discussion, scan the [WeChat QR code](https://raw.githubusercontent.com/dexmal/opendm/main/docs/image/wechat.jpeg) to contact us.

We will continue to release more model weights, technical documentation, and examples. If this project is helpful to you, please consider giving us a star on GitHub [![GitHub](https://img.shields.io/github/stars/dexmal/opendm?color=5B5BD6)](https://github.com/dexmal/opendm). Your support helps us move forward.

## Citation

```bibtex
@misc{dm05,
    title  = {{DM0.5}: An Open-World Foundation Model for General-Purpose Embodied Intelligence},
    author = {{Dexmal Team}},
    month  = {July},
    year   = {2026},
    url    = {https://www.dexmal.com/blog/dm0.5/index_en.html}
}

@misc{zhang2025vlaarena,
    title         = {VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Models},
    author        = {Borong Zhang and Jiahao Li and Jiachen Shen and Yishuai Cai and Yuhao Zhang and Yuanpei Chen and Juntao Dai and Jiaming Ji and Yaodong Yang},
    year          = {2025},
    eprint        = {2512.22539},
    archivePrefix = {arXiv},
    primaryClass  = {cs.RO},
    url           = {https://arxiv.org/abs/2512.22539},
}
```