Add model card and metadata

Hi! I'm Niels, part of the community science team at Hugging Face. I noticed this repository was missing a model card. This PR adds a README with:
- Metadata for the `robotics` pipeline tag and `transformers` library name.
- Links to the research paper [DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models](https://huggingface.co/papers/2511.15669).
- A link to the official GitHub repository.
- A summary of the model's architecture and performance results on benchmarks like LIBERO.

This will help users find and understand your work more easily on the Hugging Face Hub!

README.md

@@ -0,0 +1,40 @@
+---
+library_name: transformers
+pipeline_tag: robotics
+base_model: physical-intelligence/pi0fast_base
+tags:
+- vision-language-action
+- chain-of-thought
+- embodied-ai
+---
+
+# DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models
+
+DeepThinkVLA is a Vision-Language-Action (VLA) model designed to enhance the reasoning capabilities of robotic agents through explicit deliberation. It refactors the policy into a 2.9B parameter hybrid decoder that generates a reasoning trace (Chain-of-Thought) before emitting action chunks.
+
+- **Paper:** [DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models](https://huggingface.co/papers/2511.15669)
+- **Repository:** [https://github.com/OpenBMB/DeepThinkVLA](https://github.com/OpenBMB/DeepThinkVLA)
+
+## Model Description
+DeepThinkVLA addresses the challenges of integrating Chain-of-Thought (CoT) into VLA models by satisfying two key conditions:
+1. **Decoding Alignment:** It uses a hybrid-attention decoder that pairs causal attention for linguistic reasoning tokens with bidirectional attention for parallel action decoding.
+2. **Causal Alignment:** The model is trained via a two-stage SFT-then-RL pipeline (using GRPO) to ensure the reasoning chain is causally linked to task success.
+
+The model is initialized from the `pi0-FAST` checkpoint and demonstrates significant performance gains on robotic manipulation benchmarks.
+
+## Performance
+- **LIBERO:** 97.0% average success rate.
+- **LIBERO-Plus:** 79.0% zero-shot robustness under distribution shifts.
+- **RoboTwin 2.0:** 59.3% success rate, exceeding prior VLA baselines by significant margins.
+
+## Citation
+If you find this work helpful, please consider citing:
+
+```bibtex
+@article{yin2025deepthinkvla,
+  title={DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models},
+  author={Yin, Cheng and Lin, Yankai and Xu, Wang and Tam, Sikyuen and Zeng, Xiangrui and Liu, Zhiyuan and Yin, Zhouping},
+  journal={arXiv preprint arXiv:2511.15669},
+  year={2025}
+}
+```