---
license: cc-by-nc-4.0
tags: [bagel, vlm-gym, sokoban, reinforcement-learning, adaptive-thinking]
---

# sokoban_adaptive_vanilla3k_rl3box_step180

BAGEL-7B-MoT Sokoban checkpoint after **180 IMP-agent RL steps**.

- arm: **canonical vanilla-3k, 3-box RL**
- initialization: [`novastar111/sokoban_adaptive_vanilla_sft3k`](https://huggingface.co/novastar111/sokoban_adaptive_vanilla_sft3k)
- training set: 2,000 mixed 3-box Sokoban boards (certified-deadlock and trivial strata)
- optimization: global batch 16, 8 rollouts per prompt, learning rate 2e-6
- reward: stop-required environment success with a -0.1 malformed-output penalty
- weights: converted BF16 EMA safetensors; optimizer/training state is not included

This repository is one point in the complete 10-step release ladder from RL10 through RL200.
Evaluation uses Sokoban-v8 q95/perseg, full model rollout, move-only actions, and stop-required
success. Load with the public BAGEL-7B-MoT base/config and use `ema.safetensors` as the model delta.