Skip to content
Robotics

SciEdu_steps_20261119_relabel_100k_pi05

robot arm control policy for educational experiments

2 min read

The model is a specialized checkpoint based on the pi05 architecture (a Vision-Language-Action model by Physical Intelligence) and adapted for the LeRobot framework (a PyTorch library for real-world robotics). It was trained on a dataset of 2,915 episodes for laboratory manipulation tasks involving chemical and physical equipment. The model takes the robot state and images from three cameras (a main camera and two wrist cameras) as input and outputs control actions. It is useful for researchers and engineers working with bi_openarm_follower robot arms in educational or research laboratories.

Key features

  • Base model: trained on lerobot/pi05_base using the LeRobot library.
  • Multimodality: processes a 16-dimensional system state and high-resolution visual data from three cameras.
  • Wide range of tasks: includes handling leaves, laboratory beakers, pipettes, microscopes, electroscopes, and reagents.
  • Ready to run: includes a training configuration for 100,000 steps using the AdamW optimizer.

What you can do

  • Run on a robot: apply the ready-made control policy to perform laboratory operations using the lerobot-rollout script.
  • Conduct experiments: automate school or student chemistry and physics experiments using a robot arm.
  • Fine-tune the model: use the checkpoint for further fine-tuning on custom datasets via lerobot-train.

Example

Running the pre-built robot control policy to perform an object relocation task:

lerobot-rollout \
  --strategy.type=base \
  --robot.type=bi_openarm_follower \
  --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=leapshared/SciEdu_steps_20261119_relabel_100k_pi05 \
  --task="put the leaf on the right side" \
  --duration=60
Read moreDetails
License: apache-2.0
Size: 4.14 billion parameters
Dataset: 2,915 episodes, 903,521 frames (30 FPS)
Training parameters: 100,000 steps, batch size 32, learning rate 2.5e-05, seed 42
Discuss Telegram 0 6

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.

    # leapshared/SciEdu_steps_20261119_relabel_100k_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 leapshared/SciEdu_steps_20261119_relabel_100k_pi05 --local-dir ~/models/SciEdu_steps_20261119_relabel_100k_pi05
    # 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

  1. paligemma-3b-pt-224family root
  2. pi05_base
  3. SciEdu_steps_20261119_relabel_100k_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

leapshared/SciEdu_steps_20261119_relabel_100k_pi05
Author
leapshared
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

60

Ask AI about this model

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

Comments