Downloads · 30 days
4
14% of all-time downloads
zimplex/pi0-base-qpos-target-abs-20260526-1651-step100000
pi0-base-qpos-target-abs-20260526-1651-step100000 is a robotics model from zimplex. Use it for the robotics task on the model card, and read the license before you ship it in a product. It is set up for lerobot. The card lists the license as mit.
Final baseline checkpoint used to produce the genesis-hr-bench evaluation numbers (eval run full202605290220, 45-task evalv1100 set). Inference-only (lerobot pretrainedmodel/ layout: weights + config + pre/post-proces…
Downloads · 30 days
4
14% of all-time downloads
All-time downloads
28
Public
Parameters
4B
8.9 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8.9 GB · 100%
How the weights are stored.
BF163.6B · 90%
From the Hugging Face model README
Final baseline checkpoint used to produce the genesis-hr-bench evaluation
numbers (eval run full_20260529_0220, 45-task eval_v1_100 set). Inference-only
(lerobot pretrained_model/ layout: weights + config + pre/post-processors).
| Base model | lerobot/pi0_base |
| Dataset | zhouqh/hrbench → genesis_hr_bench_lerobot_qpos_target_abs_20260526_1651 |
| Action space | qpos_abs (8-D Franka joint absolute + gripper) |
| Step | 100,000 |
| Wandb | pi0_genesis_hr_bench/irbbsucj |
| Uploaded | 2026-08-05 07:00:01 UTC |
Note. Round-3 run on qpos_target_abs_20260526_1651. Camera-rename maps the dataset's 2 cameras onto 2 of pi0_base's 3 slots; the 3rd is zero-masked.
from lerobot.policies.factory import make_policy_from_pretrained
policy = make_policy_from_pretrained("zimplex/pi0-base-qpos-target-abs-20260526-1651-step100000", device="cuda")
policy.eval()
# Observation keys: observation.images.base_0_rgb / left_wrist_0_rgb /
# right_wrist_0_rgb (as available), observation.state. Action space: qpos_abs.
action_chunk = policy.predict_action(batch_observation)
Pair the checkpoint with the matching
qpos_absdataset variant — the normalizers andaction_dimare baked in.