Downloads · 30 days
0
g1-walk-200m
ssuleiman/g1-walk-200m
g1-walk-200m is a reinforcement learning model from ssuleiman. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for brax.
Saved locomotion policy parameters for AXIS Cortex.
Access
Public
Updated Jun 14, 2026
Repo size
37.3 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other35.8 MB · 95%
At a glance
- Task
- Reinforcement Learning
- Library
- brax
- Access
- Public
- Created
- Jun 14, 2026
- Updated
- Jun 14, 2026
- SHA
- f183aa03
Model card
From the Hugging Face model README
AXIS G1 Joystick Policy
Saved locomotion policy parameters for AXIS Cortex.
Metadata
- Robot:
g1 - Terrain:
flat - Training preset:
cloud - Network:
large/default - Format:
brax.io.model.save_params - Policy file:
policy.params
Local Evaluation
Place this file in the AXIS Cortex repository and run:
uv run python -m scripts.interactive_eval_policy policy.params --robot g1 --terrain flat --command 0.0 0.0 0.0
If this was uploaded with periodic checkpoints, they are under
checkpoints/.
- Task
- Reinforcement Learning
- Library
- brax
- Created
- Jun 14, 2026
- Updated
- Jun 14, 2026