molmoact2_insert_candle_quantile_norm_fix_25k

Fine-tuned MolmoAct2 (action-expert-only) for insert_candle on SO101 data.

Policy	MolmoAct2 (policy.type=molmoact2)
Init checkpoint	allenai/MolmoAct2
Dataset	villekuosmanen/armnetbench_insert_candle
Task	insert_candle
Action dim	12 (bimanual)
Cameras	top, left_wrist, right_wrist
Training	25k steps, QUANTILES norm, freeze, batch 32 global, Isambard GH200
Prior HF repo	pravsels/molmoact2_insert_candle
W&B project	molmoact2_insert_candle_quantile_norm_fix_25k
W&B run	b07rlk7i

Checkpoints

The checkpoint (local step 025000, 25k training steps) lives at the repository root for direct loading.

Verification

| Checkpoint step | 025000 | | Source path | checkpoints/025000/pretrained_model/ | | model.safetensors | 10,884,573,720 bytes, sha256 af86903eebd5c089c4e74c694db6345671569d38e23da12840159744e8f4b593 | | policy_preprocessor.json | 2,523 bytes, sha256 aac2ee4e61c26a5322c9b3e2f727ce060c90071e012185d9f39cceb2d43ea04a | | policy_postprocessor.json | 758 bytes, sha256 8690a8e7015281571c9de7d88073b302cd03123e5f677ea582c669dbf014e7ad | | train_config.json | 8,836 bytes, sha256 2ac88bc6f39c14eb153b6dd8a59e6686a51e1d2c2647206573d8dde433c82d1f |

Verify after download:

sha256sum model.safetensors
# expected: af86903eebd5c089c4e74c694db6345671569d38e23da12840159744e8f4b593

Usage

from lerobot.policies.molmoact2.modeling_molmoact2 import MolmoAct2Policy
policy = MolmoAct2Policy.from_pretrained("pravsels/molmoact2_insert_candle_quantile_norm_fix_25k")