Skip to content
Robotics

microduck-tug-p10

reinforcement learning policy for robot

1 min readupdated

The microduck-tug-p10 model is a perpetual policy (behavior algorithm) for the microduck robot platform, built for tug-of-war tasks. It processes 61-dimensional observations and outputs 14 control actions at 50 Hz, enabling the robot to hold its position or pull until a stop command is issued. Trained via Reinforcement Learning in the Mjlab environment using mjlab-microduck 0.1.0, it is useful for robotics researchers.

Key features

  • Perpetual movement policy running at 50 Hz until a stop command is received.
  • Built-in observation normalizer right inside the policy.onnx file for working with raw data.
  • Trained via reinforcement learning using the mjlab-microduck 0.1.0 environment (checkpoint 1840).

What you can do

  • Upload the policy to a compatible microduck robot using the robotctl utility to participate in a competition.
  • Reproduce model training locally using the microduck-challenges repository and the uv package manager.

Example

sudo robotctl policy load  SSH111111/microduck-tug-p10
Read moreDetails
Task ID: Mjlab-TugOfWar-MicroDuck
Architecture: 61-D observation, 14 actions, 50 Hz
Base: mjlab-microduck 0.1.0
Training: seed 42, checkpoint 1840
Discuss Telegram 0 31

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 0.0 GB · ready files0.5 GB
  • FP8 / INT8recommendedweights 0.0 GB0.5 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 · BF16 — 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.

    # SSH111111/microduck-tug-p10
    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 SSH111111/microduck-tug-p10 --local-dir ~/models/microduck-tug-p10
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to microduck-tug-p10
  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 microduck-tug-p10 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

SSH111111/microduck-tug-p10
Author
SSH111111
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

310

Ask AI about this model

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

Comments