Skip to content
Robotics

ACT policy for robotic manipulator

2 min read

Trained using the ACT method (Action Chunking with Transformers — an imitation learning method that predicts short chunks of actions instead of single steps), this model is built for the so-called so_follower robotic arm and designed to handle object manipulation tasks such as picking up cubes and placing them in a tray or stacking them on top of each other. The policy is based on teleoperation data and utilizes visual observations from a wrist camera. This solution is useful for robotics developers and researchers using the LeRobot ecosystem.

Key features

  • Trained and deployed using the modern LeRobot library.
  • Uses visual data from a wrist camera (observation.images.wrist) at 640x480 resolution.
  • Predicts 6-dimensional vectorized actions (action) based on state and images.
  • Built on a dataset containing recorded episodes of grasping and moving objects.

What you can do

  • Run the policy on a real so_follower robotic arm for automated cube sorting.
  • Deploy the model for autonomous execution of object grasping and stacking tasks.
  • Fine-tune your own policy using the LeRobot framework and collected demonstration data.

Example

lerobot-rollout \
  --strategy.type=base \
  --robot.type=so_follower \
  --robot.port= \
  --robot.cameras="{ : {type: opencv, index_or_path: , width: 640, height: 480, fps: 30}}" \
  --policy.path=stack \
  --task="Pick up the cube and place it in the tray" \
  --duration=60
Read moreDetails
Benchmarks: success data not provided yet
Training: 100,000 steps, batch size 64, AdamW optimizer, learning rate 1e-05
Dataset: 82 episodes, 196,610 frames at 30 FPS
Discuss Telegram 0

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 0.2 GB · ready files0.7 GB
  • BF16weights 0.1 GB0.6 GB
  • FP8 / INT8recommendedweights 0.0 GB0.6 GB
  • 4-bitweights 0.0 GB0.5 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 2060

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.

    # pschwartz26/stack
    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 pschwartz26/stack --local-dir ~/models/stack
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to stack
  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 stack (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

pschwartz26/stack
Author
pschwartz26
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