Skip to content
Robotics

TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA

Robotic Arm Control Model

2 min readupdated

A new vision-language-action model built on the pi05 architecture by Physical Intelligence for the LeRobot robotics platform. It is trained to control a bimanual robot for a complex manipulation task: picking up a cup with the right hand, moving it to the center, and placing it into a box with the left hand. The model is adapted for Fairino-type arms and processes data from three security cameras along with the robot's current state. This is a useful tool for engineers and researchers working on practical robot training using imitation learning.

Key features

  • Specialization: Trained on a precise two-handed object transfer task using Fairino robots.
  • Multimodality: Processes visual data from three cameras (top, left, right) and system states.
  • LeRobot framework: Full integration with the Hugging Face LeRobot ecosystem for training and inference.
  • Training: The model was trained on a dataset of 75 episodes (49,128 frames) using the AdamW optimizer.

What you can do

  • Run on a robot: Deploy the ready-made policy on a real Fairino robot to automate assembly or object transfer.
  • Data analysis: Explore the demonstration dataset and movement structure in the training episodes via the LeRobot visualizer.
  • Fine-tuning: Use the lerobot/pi05_base base model to create custom robotic skills.

Example

Command to run the policy on a real robot:

lerobot-rollout \
  --strategy.type=base \
  --robot.type=Fairino \
  --robot.port= \
  --robot.cameras="{ : {type: opencv, index_or_path: , width: 640, height: 480, fps: 30}, : {type: opencv, index_or_path: , width: 640, height: 480, fps: 30}}" \
  --policy.path=Bekhzod/TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA \
  --task="pick up the cube with the right arm, place it in the middle, then put it in the box with the left arm." \
  --duration=60
Read moreDetails
License: apache-2.0
Base model: lerobot/pi05_base
Training configuration: 20,000 steps, batch size 32, learning rate 0.001, seed 1000
Input parameters: state (14,) and images from three cameras (3, 480, 640)
Output parameters: action vector (14,)
Discuss Telegram 0 21

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 13 GB · ready files15 GB
  • BF16weights 6.8 GB7.9 GB
  • FP8 / INT8recommendedweights 3.4 GB4.2 GB
  • 4-bitweights 1.9 GB2.6 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · F32 — from RTX 5080

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

This is a LoRA: counted together with its base model pi05_base.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

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.

    # Bekhzod/TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    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 lerobot/pi05_base --local-dir ~/models/pi05_base
    hf download Bekhzod/TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA --local-dir ~/models/TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to pi05_base
  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

Requires the base model

lerobot/pi05_base

This model is a LoRA / adapter. Download the base model first, then load this one on top.

  1. paligemma-3b-pt-224family root
  2. pi05_base
  3. TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA

New posts about versions of paligemma-3b-pt-224 (quants, finetunes, LoRAs) will show up under the bell.

Fine-tunes 12All on Hugging Face
LoRAs and adapters 12All on Hugging Face

About the model

Bekhzod/TASK_01_bimanual_cube_transfer_teleop_pi05_LoRA
Author
Bekhzod
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

210

Ask AI about this model

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

Comments