InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulation

InternVLA-A1 integrates understanding, generation, and action experts via a Mixture-of-Transformers (MoT) framework, which synergizes MLLMs' semantic reasoning with world-model-style dynamics prediction to guide action execution.

Building upon InternVL3 and Qwen3-VL, we instantiate InternVLA-A1 at 2B and 3B parameter scales. Covering different model scales and pre-training data configurations, we release the InternVLA-A1 series:

• InternVLA-A1-3B: pretrained on heterogeneous data sources over real-robot data, synthetic simulation data, and human videos
• InternVLA-A1-3B-RoboTwin: finetuned on RoboTwin 2.0 benchmark
• InternVLA-A1-3B-Pretrain-InternData-A1: pretrained on InternData-A1 only
• InternVLA-A1-2B-Pretrain-InternData-A1: pretrained on InternData-A1 only

Evaluation on RoboTwin 2.0 Simulation Benchmark

Setting: All models are jointly fine-tuned across 50 tasks (50 clean + 500 randomized demos each).

Performance Summary: InternVLA-A1-3B achieves the highest success rates across both Easy and Hard settings on the RoboTwin 2.0 Benchmark (averaged over 50 tasks).

🔑 Key Features

• 🔮 The Core: Synergizes MLLM's semantic understanding with world-model-style dynamic prediction, enabling it to "imagine" the future and guide adaptive actions.
• 🚀 The Fuel: Enables joint training on heterogeneous data sources over real-world robot data, synthetic simulation data, and egocentric human videos.
• ⚡ The Output: Tackles highly dynamic scenarios with effortless mastery.

Usage

Please refer to our official repo InternVLA-A1.

Demonstrations

InternVLA-A1 exhibits consistent robustness across static manipulation, dynamic manipulation, and simulation benchmarks, especially demonstrating remarkable superiority in dynamic scenarios.

License and Citation

All the code within this repo are under CC BY-NC-SA 4.0. Please consider citing our project if it helps your research.

@article{internvla_a1,
  title={InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulation},
  author={Cai, Junhao and Cai, Zetao and Cao, Jiafei and Chen, Yilun and He, Zeyu and Jiang, Lei and Li, Hang and Li, Hengjie and Li, Yang and Liu, Yufei and others},
  journal={arXiv preprint arXiv:2601.02456},
  year={2026}
}

Acknowledgments

• Lerobot
• openpi
• InternVL
• Qwen3-VL
• COSMOS