Skip to content
Robotics

smolvla pill_sorting_colour_cf_swap_mix_ft_v2

политика сортировки таблеток для so_follower

1 min read

Это компактная vision-language-action политика SmolVLA, дообученная TomasAnderegg в LeRobot для робота so_follower. Модель получает состояние и изображение с фронтальной камеры и выдаёт команды действий; по метаданным — около 0,45 млрд параметров.

Ключевые особенности

  • Compact VLA-политика на базе SmolVLA, рассчитанная на consumer-grade железо.
  • Вход: observation.state формы (6,) и изображение observation.images.front (3, 480, 640).
  • Выход: action формы (6,).
  • Обучена на 116 эпизодах / 52 064 кадрах при 30 FPS в датасете pill_sorting_colour_cf_swap_mix_v2.
  • 16 задач: разложить синюю, красную, жёлтую или зелёную таблетку в квадратное, круглое, треугольное или крестовое отделение.

Пример

lerobot-rollout --strategy.type=base --robot.type=so_follower --robot.port=<your_robot_port> --robot.cameras="{ front: {type: opencv, index_or_path: <index>, width: 640, height: 480, fps: 30}}" --policy.path=TomasAnderegg/pill_sorting_colour_cf_swap_mix_ft_v2 --task="Put the blue pill in the square compartment." --duration=60
ПодробнееЛицензия Apache-2.0; дообучение от TomasAnderegg/pill_sorting_full_ft_a100_v3_dagger60.
Обучение: 4000 шагов, batch size 64, AdamW, learning rate 1e-5, seed 1000.
Робот so_follower, одна камера front; данные о требованиях к VRAM в карточке не указаны.
Опубликованных результатов оценки пока нет.
Discuss Telegram 0 26

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.

    # TomasAnderegg/pill_sorting_colour_cf_swap_mix_ft_v2
    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 TomasAnderegg/pill_sorting_colour_cf_swap_mix_ft_v2 --local-dir ~/models/pill_sorting_colour_cf_swap_mix_ft_v2
    # 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 pill_sorting_colour_cf_swap_mix_ft_v2 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

TomasAnderegg/pill_sorting_colour_cf_swap_mix_ft_v2
Author
TomasAnderegg
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

260

Ask AI about this model

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

Comments