Downloads · 30 days
0
ankile/openpi-pi05-01h-a4-marker-ours-freecf-20k-idle-ft-iris
openpi-pi05-01h-a4-marker-ours-freecf-20k-idle-ft-iris is a robotics model from ankile. 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 openpi. The card lists the license as apache-2.0.
A Pi0.5 model fine-tuned using the OpenPI framework.
Downloads · 30 days
0
Access
Public
Updated Jun 14, 2026
Repo size
5.2 GB
Likes
0
Public
Click a slice to open those files.
Other5.2 GB · 100%
From the Hugging Face model README
A Pi0.5 model fine-tuned using the OpenPI framework.
| Property | Value |
|---|---|
| OpenPI config | pi05_sir_droid_finetune_idle |
| Checkpoint step | 19999 |
| Training data | N/A |
| Precision | bfloat16 |
| Parameter size | ~6.7 GB |
| Source checkpoint | /iris/u/ankile/openpi-pi05-real01c/checkpoints/pi05_sir_droid_finetune_idle/sir_01h_a4_marker_ours_freecf_20k_bs32_idle_iris/19999 |
| Hugging Face repo | ankile/openpi-pi05-01h-a4-marker-ours-freecf-20k-idle-ft-iris |
| W&B run | link |
| SLURM job ID | 15890575 |
# Download checkpoint from HF Hub
huggingface-cli download ankile/openpi-pi05-01h-a4-marker-ours-freecf-20k-idle-ft-iris --local-dir <local_path>
# Run inference server
cd deps/openpi
uv run python scripts/serve_policy.py pi05_sir_droid_finetune_idle \
--checkpoint-dir <local_path>
from openpi.training import config as openpi_config
from openpi.policies import policy_config as openpi_policy_config
train_config = openpi_config.get_config("pi05_sir_droid_finetune_idle")
policy = openpi_policy_config.create_trained_policy(
train_config, "<local_path>"
)
result = policy.infer(obs_dict)
actions = result["actions"]
Orbax format, all parameters in bfloat16.
├── _CHECKPOINT_METADATA
├── checkpoint_provenance.json
├── openpi_config.json
├── assets/
│ └── (normalization stats)
└── params/
└── (orbax checkpoint files)
Apache 2.0