---
license: apache-2.0
tags:
  - robotics
  - vla
  - pi0
  - droid
---

# MolmoBot-Pi0-DROID-absjoint-reasoninit

MolmoBot-Pi0 (PaliGemma-decoder VLA) **full fine-tune** on the equad DROID data
(`shrg7/droid-equad-stage-1`), with **absolute joint-pose** action targets
(`arm[t] = q[t+1]`, 7 joints + gripper).

**Reasoning-init variant:** before full fine-tuning, a frozen reasoning LoRA
(trained reasoning-only on the same DROID data) is folded into the PaliGemma
decoder weights in-place, so the full-FT starts from a reasoning-initialized
backbone (a weight-space prior), rather than plain base weights.

- Config: `molmobot_pi0_lerobot_droid_absjoint`
- **Checkpoint: step 2000 (intermediate — training in progress)**, held-out
  `val_action_loss = 0.0199`.
- Contents: `model.safetensors` (flat, full model), `metadata.pt`,
  `assets/droid_equad/norm_stats.json`, `assets/train_config.pkl`.
  Optimizer state omitted.