---
tags:
- framework:openpi
- model:pi05
- platform:trossen-ai
- robot:stationary
- task:block-transfer
---
# pi05-block-transfer-trossen-ai-openpi

A **π₀.₅ (pi0.5)** policy fine-tuned with **openpi** (Physical Intelligence's JAX trainer) for the **block transfer** task on the **Trossen AI Stationary** bimanual platform.

> **Framework:** openpi (JAX/XLA) — this is **not** a LeRobot/PyTorch checkpoint. Load it with the openpi runtime, not `lerobot`.

## Inputs

- **Dataset:** [`TrossenRoboticsCommunity/trossen_ai_stationary_block_transfer`](https://huggingface.co/datasets/TrossenRoboticsCommunity/trossen_ai_stationary_block_transfer) — LeRobot **v2** format (openpi requires v2; v3 is rejected).
- **Base model:** `π₀.₅ (pi0.5)` pretrained checkpoint from Physical Intelligence, LoRA fine-tuned.
- **Training code:** [`TrossenRobotics/openpi`](https://github.com/TrossenRobotics/openpi)
- **Hardware:** Trossen AI Stationary (bimanual), trained locally on an **RTX 5090**.
- **Action/state space:** 14-dim (per `assets/trossen/norm_stats.json`).

## Configuration

- **Steps:** 100,000
- **Batch size:** 8
- **Fine-tuning:** LoRA
- Norm stats: `assets/trossen/norm_stats.json` (auto-computed at train time).

## How it was trained

Fine-tuned locally on an RTX 5090 using the openpi JAX trainer (LoRA, batch size 8, 100K steps). openpi's first-class π implementation generally outperforms LeRobot's PyTorch π policies. Exported as the final orbax checkpoint (`params/` = weights, `train_state/` = optimizer state for resuming).

## How to use

```bash
git clone https://github.com/TrossenRobotics/openpi
# follow openpi serving/inference instructions; point the policy at this repo's params/
# recommended inference rate: 50 Hz
```

## Outputs

- **Best checkpoint:** step 100,000
- **Checkpoint exported:** 2025-10-29

## Evaluation

<!-- Upload eval clips to videos/ in this repo; this player renders them automatically. -->
<video controls width="640" src="https://huggingface.co/TrossenRoboticsCommunity/pi05-block-transfer-trossen-ai-openpi/resolve/main/videos/eval.mp4"></video>

## Versions

- r1 (2025-10-29) — initial openpi checkpoint export.