Downloads · 30 days
0
meishigan55/genie-libero-policy
genie-libero-policy is a machine learning model from meishigan55. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for diffusers. The card lists the license as apache-2.0.
--- license: apache-2.0 tags: - robotics - vision-language-action - policy-learning - libero - genie-envisioner datasets: - libero pipelinetag: robotics ---
Downloads · 30 days
0
Access
Public
Updated Dec 26, 2025
Repo size
16.9 GB
Likes
0
Public
Click a slice to open those files.
.safetensors16.9 GB · 100%
From the Hugging Face model README
license: apache-2.0 tags:
This repository contains policy adaptation checkpoints for Genie-Envisioner trained on the LIBERO benchmark.
Genie-Envisioner is a world model-based approach for robot manipulation that learns to predict future visual observations and uses these predictions to guide policy learning. These checkpoints are the policy models adapted from the pre-trained video prediction model.
| Checkpoint | Task Suite | Description |
|---|---|---|
policy_adaptation_libero_spatial | LIBERO-Spatial | 10 tasks focusing on spatial relationships |
policy_adaptation_libero_object | LIBERO-Object | 10 tasks focusing on object manipulation |
policy_adaptation_libero_goal | LIBERO-Goal | 10 tasks with diverse goals |
policy_adaptation_libero_10 | LIBERO-Long | 10 long-horizon tasks |