---
language:
- en
library_name: lerobot
base_model: allenai/MolmoAct2
datasets:
- Spa-Bench/spa-bench-training-teleoperation-1200
pipeline_tag: robotics
tags:
- robotics
- vision-language-action
- so-101
- spa-bench
---

# MolmoAct2 — Spa-Bench Step 76,596

This checkpoint is released as an anonymous supplementary artifact for a paper under double-blind review. It was evaluated on a physical SO-101 robot in Spa-Bench.

## Model details

| Field | Value |
| --- | --- |
| Checkpoint | End of epoch 12; step 76,596 |
| Inputs | Middle RGB, wrist RGB, six joint positions, and a text instruction |
| Outputs | Six absolute joint-position targets |
| Action horizon | 30 |
| Adaptation | VLM and continuous-action expert updated; token embeddings frozen |
| Optimizer | AdamW; zero weight decay; component-specific learning rates |

Training-data documentation: [`spa-bench-training-teleoperation-1200`](https://huggingface.co/datasets/Spa-Bench/spa-bench-training-teleoperation-1200).

## Physical evaluation

87/120 familiar trials, 156/300 withheld-composition trials, 73/120 matched withheld trials, and 107/120 matched direct-manipulation controls. These are physical rollout results, not simulation metrics.

Rollouts: [`spa-bench-eval-rollouts-molmoact2-full`](https://huggingface.co/datasets/Spa-Bench/spa-bench-eval-rollouts-molmoact2-full).

## Limitations and safety

Results apply to this checkpoint, SO-101 embodiment, workspace, cameras, objects, and physical protocol. Robot policies can move hardware unexpectedly. Use conservative limits, an accessible emergency stop, a clear workspace, and direct supervision. Do not deploy unattended or in safety-critical settings.

## Double-blind release note

Author, institution, source-repository, and archival citation details are intentionally omitted during review. They will be restored in the archival release.