Skip to content
Robotics

microduck-sprint2m-scale2048

непрерывная походка для microduck

1 min read

Политика ходьбы для робота microduck от atul-shukla2072. Принимает 61-мерные наблюдения, выдаёт 14 действий и работает на частоте 50 Гц. Это не разовый маневр, а постоянная политика: загружается в слот и работает, пока её не остановят.

Ключевые особенности

  • 61-D наблюдений и 14 действий — формат под microduck.
  • Политика рассчитана на непрерывную работу до явной остановки.
  • Нормализатор наблюдений встроен в policy.onnx: на вход можно подавать сырые наблюдения.
  • manifest.json соответствует schema 2 манифеста политик microduck.
  • Обучена в задаче Mjlab-Sprint2m-MicroDuck, checkpoint 999, seed 1.

Пример

sudo robotctl policy load <slot> atul-shukla2072/microduck-sprint2m-scale2048
ПодробнееОбучение шло с 2048 параллельными средами и до 1000 итераций, seed 1, checkpoint 999.
Повторный запуск с тем же кодом, uv.lock, командой и seed даёт сопоставимую политику, но не те же веса: GPU RL не бит-репродуцируем между машинами.
Требования к VRAM в карточке не указаны.
Discuss Telegram 1 14

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

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.

    # atul-shukla2072/microduck-sprint2m-scale2048
    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 atul-shukla2072/microduck-sprint2m-scale2048 --local-dir ~/models/microduck-sprint2m-scale2048
    # 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

New posts about versions of microduck-sprint2m-scale2048 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

atul-shukla2072/microduck-sprint2m-scale2048
Author
atul-shukla2072
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

141

Ask AI about this model

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

Comments