Skip to content
Robotics

pi05_telearms_v5_exp_score_1500

полный файнтюн π₀.₅ для двуручной манипуляции

1 min read

Это чекпоинт от helen9975 на базе lerobot/pi05_base. Модель полностью дообучена для симулированных двуручных задач YAM: 10 задач и три камеры 224×224. Размер — около 4,14 млрд параметров.

Ключевые особенности

  • Полный fine-tune LLM, vision и action expert; bf16 и gradient checkpointing.
  • 15 000 симулированных демонстраций YAM, 9 115 598 кадров при 30 Hz; без domain randomization.
  • По 1 500 демонстраций на задачу из band высокой proficiency по demo score.
  • 30 000 шагов и 15,36 млн samples; 8×H100, глобальный batch 512.
  • Один из восьми v5-экспериментов с одинаковым рецептом и бюджетом для сравнения.

Пример

Сценарий из карточки: политике подают три потока с камер 224×224 и вариант промпта задачи; она выдаёт 50-шаговые action chunks. Действия относительные, gripper — абсолютные. Для запуска нужен LeRobot 0.6.1 с патчем репозитория; файлы лежат в pretrained_model/.

ПодробнееTraining loss: 0,018.
Обучение: AdamW, peak LR 1e-4, 1 000 warmup, cosine decay до 1e-5.
Ограничение: в симуляции ещё не оценивалась; validation loss нет, так как эпизоды не откладывались.
Чекпоинт — step 30 000.
Discuss Telegram 0 32

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.

    # helen9975/pi05_telearms_v5_exp_score_1500
    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 helen9975/pi05_telearms_v5_exp_score_1500 --local-dir ~/models/pi05_telearms_v5_exp_score_1500
    # 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 pi05_telearms_v5_exp_score_1500 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

helen9975/pi05_telearms_v5_exp_score_1500
Author
helen9975
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

320

Ask AI about this model

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

Comments