Skip to content
Robotics

act-green-square-test

robot control model for green square selection

2 min read

A manipulation control policy trained using the LeRobot framework and the ACT method (Action Chunking with Transformers — an imitation learning approach that predicts short action sequences instead of single steps). The model takes the robot state and images from two cameras (front and gripper) at a 640x480 resolution as input, and then outputs control vector commands to select the target object. The solution is built for so_follower type robots based on a dataset of 29 episodes and will be useful for robotics developers using the LeRobot ecosystem.

Key features

  • ACT architecture: prediction of short action sequences instead of step-by-step control
  • Multimodal data processing: joint states and two camera streams (front and gripper)
  • Trained on a dataset of 29 episodes (14,236 frames at 30 FPS)
  • Integration with LeRobot library version 0.7.1 and the adamw optimizer

What you can do

  • Run the ready-made control policy on a real so_follower robot using the lerobot-rollout command
  • Train a custom policy on user movement demonstrations using LeRobot scripts
  • Configure camera parameters and connection ports for a specific robot hardware setup

Example

lerobot-rollout \
  --strategy.type=base \
  --robot.type=so_follower \
  --robot.port= \
  --robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 30}, gripper: {type: opencv, index_or_path: 1, width: 640, height: 480, fps: 30}}" \
  --policy.path=albarodriguezsalvador/act-green-square-test \
  --task="green-square-fix-movement-001 selects the green square between 4 squares"
Read moreDetails
Training parameters: 200 steps, batch size 8, learning rate 1e-05, seed 1000.
Input features: observation.state of dimension (6,), front and gripper images of dimension (3, 480, 640).
Outputs: action vector of dimension (6,).
Evaluation results on a real robot have not yet been provided by the author.
Discuss Telegram 0 17

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.

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

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

About the model

albarodriguezsalvador/act-green-square-test
Author
albarodriguezsalvador
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

170

Ask AI about this model

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

Comments