Skip to content
Robotics

act_sos101_test_50k

robotics control policy for robotic arms

2 min readupdated

Trained with the LeRobot library, this control policy implements the ACT method (Action Chunking with Transformers—an imitation learning approach that predicts short action sequences instead of single steps). The model controls an so_follower robotic arm using visual input from a front camera and current joint states. It was trained on teleoperation demonstrations to pick up a red ball and place it into a black basket, making it useful for researchers and engineers working with accessible LeRobot-based hardware.

Key features

  • Hardware compatibility: designed for so_follower robotic arms.
  • Visual input: processes front-camera feeds at 480x640 resolution.
  • ACT method: predicts action chunks for smooth, accurate task execution.
  • Task focus: trained specifically to grasp a red ball and move it into a black basket.

What you can do

  • Deploy on hardware: run the pretrained policy to control a compatible robotic arm in real time.
  • Fine-tune: use the 52-episode dataset to train or customize your own control policies.
  • Analyze trajectories: inspect the dataset in the LeRobot visualizer to study motion paths.

Example

Run the pretrained control policy on your robot:

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=googaar/act_sos101_test_50k \
  --task="Put red ball in black basket" \
  --duration=60
Read moreDetails
Parameters: 0.05B
Training: 50,000 steps, batch size 8, adamw optimizer, learning rate 1e-05
Dataset: 52 episodes, 29,867 frames at 30 FPS
Discuss Telegram 0 14

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.

    # googaar/act_sos101_test_50k
    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 googaar/act_sos101_test_50k --local-dir ~/models/act_sos101_test_50k
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to act_sos101_test_50k
  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_sos101_test_50k (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

googaar/act_sos101_test_50k
Author
googaar
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

140

Ask AI about this model

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

Comments

Robotics

HINT-WAM

HINT-WAM