You can deploy this model and start earning money today!

vladimirmodylevskii/
tomato-harvest-so100-checkpoint-1000

Name: tomato-harvest-so100-checkpoint-1000
Availability: InStock

safetensors

Gr00tN1d7

Views

No views yet

GR00T N1.7 SO-100 Tomato Harvest — Checkpoint 1000

Fine-tuned checkpoint from NVIDIA GR00T-N1.7-3B on the SO-100 tomato harvest task.

Training

Dataset: vladimirmodylevskii/tomato-harvest-final-v1-clean
Embodiment tag: NEW_EMBODIMENT
Cameras: wrist (front) + top
Task: Pick the ripe tomato and place it in the bowl
Global step: 1000 / 10000
Batch size: 32

Usage

python

from gr00t.policy.gr00t_policy import Gr00tPolicy

policy = Gr00tPolicy(
    embodiment_tag="NEW_EMBODIMENT",
    model_path="vladimirmodylevskii/tomato-harvest-so100-checkpoint-1000",
    device="cuda:0",
)

Or download locally and point model_path to the checkpoint directory.