README.md

---
library_name: opentau
tags:
- robotics
- vla
- pi0
- libero
- Reinforcement Learning
- manipulation
- flow-matching
- pytorch
license: apache-2.0
datasets:
- physical-intelligence/libero
repo_url: https://github.com/TensorAuto/OpenTau
---

# moka_pot_RECAP_R0

A **pi0 (π₀) RECAP** Vision-Language-Action (VLA) model, finetuned on the **LIBERO** robotic manipulation benchmark using the **OpenTau** training framework. This model is designed to follow natural language instructions to perform manipulation tasks in a simulated tabletop environment.
Achieves **~89% success rate** measured over **320 episodes**.

**For full documentation, evaluation results, and inference code, please visit the repository:**
<br>
👉 **[https://github.com/TensorAuto/OpenTau](https://github.com/TensorAuto/OpenTau)**

---

## Model Details

### Description
- **Model Type:** Vision-Language-Action (VLA) Model
- **Base Architecture:** π₀ (pi0) by Physical Intelligence
- **Backbone:** PaliGemma-3B (VLM) + Gemma-300M (Action Expert) + RL indicator
- **Training Data:** Moka Pot Task on LIBERO (Lifelong Robot Learning) Benchmark
- **Framework:** OpenTau

### Architecture
The **PI0 RECAP** architecture uses a flow-matching and Reinforcement Learning policy designed for open-world generalization. It combines a Visual Language Model (VLM) for high-level semantic understanding with a smaller "action expert" model that generates continuous joint trajectories (10-step action chunks) via flow matching. It uses RL to learn from good and bad episodes

---

## Training and Evaluation
The Advantage Indicator (It) was set to True for only 10% of datapoints. 

### Dataset
This model was finetuned on the **Moka Pot task in LIBERO 10** benchmark dataset and autonomous rollouts. It consists of around 29 expert teleoperated episodes and 212 autonomous rollouts under moka_pot_libero_sft policy.

### Results
For detailed usage instructions, success rates, baseline comparisons, and evaluation protocols, please refer to the [OpenTau GitHub Repository](https://github.com/TensorAuto/OpenTau).
Achieves **~89% success rate** measured over **320 episodes**.