Skip to content
Robotics

gaze_act

политика LeRobot для тостов

1 min read

gaze_act — это policy на 0.05B параметров от fecasado, обученная в LeRobot для робота blueberry_ros. Она выполняет две задачи: достать ломтик хлеба из тостера и положить его на тарелку или поверх сэндвича. На вход подаются изображения с камер right, left, user и 55-мерное состояние; выход — 26-мерный вектор действия.

Ключевые особенности

  • Обучена под конкретного робота blueberry_ros; в карточке также указана камера user_gaze.
  • Вход: три изображения 240×320 и observation.state размерности 55.
  • Выход: action размерности 26.
  • Датасет: 115 эпизодов, 75 032 кадра, 15 FPS; две задачи с тостами.
  • Обучение: 100 000 шагов, batch 128, AdamW, lr 1e-5, seed 1000, LeRobot 0.6.1.
  • Оценки на реальном роботе в карточке нет.

Пример

lerobot-rollout \
  --strategy.type=base \
  --robot.type=blueberry_ros \
  --robot.port=<your_robot_port> \
  --policy.path=fecasado/gact-kitchen-toasts-multi \
  --task="Take one slice of bread out of the toaster and place it on the empty plate" \
  --duration=60
ПодробнееБенчмарков нет: в карточке указано «No evaluation results».
Данные: 115 эпизодов, 75 032 кадра, 15 FPS.
Обучение: 100 000 шагов, batch 128, AdamW, lr 1e-5, seed 1000, LeRobot 0.6.1.
Ограничения: политика заточена под blueberry_ros и конкретные камеры; переносимость не проверена.
Discuss Telegram 0 37

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.

    # fecasado/gact-kitchen-toasts-multi
    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 fecasado/gact-kitchen-toasts-multi --local-dir ~/models/gact-kitchen-toasts-multi
    # 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 gact-kitchen-toasts-multi (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

fecasado/gact-kitchen-toasts-multi
Author
fecasado
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

370

Ask AI about this model

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

Comments