Skip to content
Robotics

BorjaBG/modelo_servir_2

Robotic policy for a manipulator arm

1 min read

Trained using ACT (Action Chunking with Transformers — an imitation learning method that predicts short action chunks instead of single steps), this model is designed for the so_follower robotic manipulator arm. It enables the task of serving a bowl to a human based on teleoperation data. Created using the LeRobot library, the model utilizes both a front-facing camera and an end-effector camera on the robot's gripper. This is useful for researchers and practitioners in home and laboratory robotics using LeRobot.

Key features

  • ACT architecture built on the LeRobot robotics framework.
  • Visual observation processing from two cameras (front and gripper-mounted).
  • Tailored for performing a specific household serving task.
  • Trained on a dataset of 30 episodes and 10,325 frames.

What you can do

  • Run the policy on a compatible so_follower robot to autonomously execute the serving task.
  • Train your own similar model on a custom dataset using LeRobot utilities.

Example

Running the ready-made robot control policy:

lerobot-rollout \
  --strategy.type=base \
  --robot.type=so_follower \
  --robot.port= \
  --policy.path=BorjaBG/modelo_servir_2 \
  --task="Servir el bol a una persona" \
  --duration=60
Read moreDetails
Training configuration: 18,000 steps, batch size 8, AdamW optimizer, learning rate 1e-05.
Hardware: 640x480 cameras at 30 FPS, 6-DOF manipulator state.
Dataset: BorjaBG/Servir (30 episodes, 10,325 frames).
Discuss Telegram 0 17

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 0.2 GB · ready files0.7 GB
  • BF16weights 0.1 GB0.6 GB
  • FP8 / INT8recommendedweights 0.0 GB0.6 GB
  • 4-bitweights 0.0 GB0.5 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 2060 · F32 — from RTX 2060

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.

    # BorjaBG/modelo_servir_2
    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 BorjaBG/modelo_servir_2 --local-dir ~/models/modelo_servir_2
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to modelo_servir_2
  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 modelo_servir_2 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

BorjaBG/modelo_servir_2
Author
BorjaBG
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

170

Ask AI about this model

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

Comments