Skip to content
Robotics

Long-WAM Sorting Dolls

чекпоинт для сортировки матрёшек

1 min read

Это evaluation-веса для задачи sorting dolls, обученные от Efficient-Large-Model/Long-WAM-YAM и сохранённые на шаге оптимизатора 20 000. Датасет — 100 демонстраций и 168 126 кадров; обучение шло на четырёх A100. Робот должен расставить пять русских матрёшек по росту слева направо.

Ключевые особенности

  • 14D absolute joint/gripper targets: по 6 суставов + gripper на каждую руку.
  • Камеры top_head, hand_left и hand_right; обработанный кадр 384×320.
  • P48 history и 32-step action chunk для управления.
  • Нормализация z-score по статистике конкретного запуска, без clipping.
  • В комплекте model.pt, config.yaml, dataset_stats.json, requirements-sip.txt и code.zip.

Пример

Инструкция для политики: “Arrange the five Russian dolls upright in a row from smallest to largest, from left to right.” На вход подаются кадры с трёх камер и 14D-цели; на выходе — абсолютные команды суставам и gripper в исходных единицах.

ПодробнееБенчмарки и success rate не заявлены. Нужны отдельные Wan2.2 VAE, UMT5 text encoder и tokenizer.
Чекпоинт не содержит полного состояния для возобновления обучения.
Обучение: global batch 48, LR 1e-5, AdamW, weight decay 0.01, BF16 / ZeRO-1, seed 42.
Лицензия базовой модели — по ссылке в карточке.
Discuss Telegram 0 18

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 11 GB · ready files13 GB
  • FP8 / INT8recommendedweights 5.6 GB6.7 GB
  • 4-bitweights 3.2 GB4.0 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 5060 Ti 8GB · BF16 — from RTX 5080

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.

    # dddddsds31341/Long-WAM-Sorting-Dolls-30k-Step-20000
    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 dddddsds31341/Long-WAM-Sorting-Dolls-30k-Step-20000 --local-dir ~/models/Long-WAM-Sorting-Dolls-30k-Step-20000
    # 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 Long-WAM-YAM (quants, finetunes, LoRAs) will show up under the bell.

Fine-tunes 12All on Hugging Face

About the model

dddddsds31341/Long-WAM-Sorting-Dolls-30k-Step-20000
Author
dddddsds31341
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

180

Ask AI about this model

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

Comments

Robotics

SmolVLA

SmolVLA