Skip to content
Robotics

pi05 lw_benchhub_pi05

ready-to-use robot control model

2 min read

The model is based on the π₀.₅ (Pi05) architecture by Physical Intelligence and is trained using the LeRobot library to control a double_piper robot. It processes visual data from multiple cameras and the robot's current state, outputting control actions to perform various kitchen and household manipulation tasks. The model is built for robotics engineers and researchers looking to apply imitation learning methods on real hardware.

Key features

  • Designed for open-world generalization in new environments.
  • Tailored for double_piper configuration robots.
  • Uses data from three cameras (base and two wrist) along with the robot's state.
  • Trained on a dataset with over 6 million frames and 6,500 episodes.

What you can do

  • Run the model to perform household tasks involving stove, tableware, and kitchen drawer control.
  • Perform inference and test the control policy on a real robot using LeRobot CLI tools.
  • Fine-tune the base model on your own collected demonstrations.

Example

Running the ready-to-use robot control policy via the command line:

lerobot-rollout \
  --strategy.type=base \
  --robot.type=double_piper \
  --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=whosricky/lw_benchhub_pi05 \
  --task="L10K3TurnOnTheStoveAndPutTheMokaPotOnIt" \
  --duration=60
Read moreDetails
Base model: lerobot/pi05_base (size around 4.14B parameters).
Optimizer: adamw with learning rate 2.5e-05, batch size: 16.
Training steps: 120,000 at a 50 FPS frame rate.
License: gemma.
Discuss Telegram 0

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 15 GB · ready files17 GB
  • BF16weights 7.7 GB9.0 GB
  • FP8 / INT8recommendedweights 3.9 GB4.7 GB
  • 4-bitweights 2.2 GB2.9 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 4090

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

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.

    # whosricky/lw_benchhub_pi05
    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 whosricky/lw_benchhub_pi05 --local-dir ~/models/lw_benchhub_pi05
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to lw_benchhub_pi05
  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

  1. paligemma-3b-pt-224family root
  2. pi05_base
  3. lw_benchhub_pi05

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

whosricky/lw_benchhub_pi05
Author
whosricky
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

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

Comments