Skip to content
Robotics

π₀.₅ cube2cross

VLA-политика для UR5e

1 min read

Это дообученная VLA-политика π₀.₅ (Pi05) от Physical Intelligence в формате LeRobot. Автор felsager дообучил lerobot/pi05_base под робота ur5e_hand_e: робот берёт цветной кубик и ставит его на красный крест. В модели 4,14 млрд параметров.

Ключевые особенности

  • Вход: состояние (10) и две камеры top/wrist 480×640; выход — action (10).
  • Дообучение на 999 эпизодах и 510 363 кадрах при 30 FPS; 4 задачи с кубиками.
  • 60 000 шагов, batch 16, AdamW, lr 5e-05; LeRobot 0.6.2.
  • Запуск через lerobot-rollout; имена камер должны совпадать с observation keys.
  • Оценки на реальном роботе в карточке нет.

Пример

lerobot-rollout \
  --strategy.type=base \
  --robot.type=ur5e_hand_e \
  --robot.port=<your_robot_port> \
  --robot.cameras='{ top: {...}, wrist: {...} }' \
  --policy.path=felsager/pi05_cube2cross_dr_bs128 \
  --task='pick up the green cube and place it on the red cross.' \
  --duration=60
ПодробнееДанные: 999 эпизодов, 510 363 кадра, 30 FPS.
Обучение: 60 000 шагов, batch 16, AdamW, lr 5e-05, seed 0, LeRobot 0.6.2.
Ограничения: только задачи с кубиками и красным крестом; робот ur5e_hand_e; камеры top и wrist.
Eval: результатов на реальном роботе нет; VRAM не указана.
Discuss Telegram 0 40

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # felsager/pi05_cube2cross_dr_bs128
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download felsager/pi05_cube2cross_dr_bs128 --local-dir ~/models/pi05_cube2cross_dr_bs128
    # The model is in ~/models. How to run it is in the model card (link above).
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of pi05_cube2cross_dr_bs128 (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

felsager/pi05_cube2cross_dr_bs128
Author
felsager
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

400

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments