Skip to content
Robotics

FLUX 3 Action EE-50

Dual-arm robotic system for towel folding

1 min readupdated

This model is a specialized version of the base FLUX 3 Action architecture (6.95 billion parameters), fine-tuned using the EE-50 recipe for the task of dual-arm towel folding based on the UMI towel folding dataset. The model outputs action sequences for robotic manipulators. It differs from the base version by its narrow specialization in the household task of soft-object manipulation and precise weight tuning. The model will be useful for robotics researchers and developers.

Key features

  • Built on the base FLUX 3 Action with 6.95 billion parameters.
  • Trained on 285 episodes of the UMI towel folding dataset.
  • Uses the EE-50 training recipe with the AdamW optimizer and Power EMA.
  • Supports inference using specialized configuration presets.

What you can do

  • Integrate the model into a robotics project to automate movements.
  • Test trajectory generation for dual-arm manipulators.
  • Use configuration files and scripts to evaluate accuracy.

Example

Example of loading the model using a method from the repository:

policy = FluxActionPolicy.from_pretrained("nishankgite/towel-fold-flux3-action-ee50")
Read moreDetails
Benchmarks: average position error k=1 from 7.5 / 7.0 mm
Hardware requirements: one B200 card, latency 244 ms p50
Training: 6000 steps, ema_0p10, bf16, peak LR 2e-4
License: FLUX Kommunity License v1.0 (non-commercial use only)
Discuss Telegram 0 30

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 13 GB · ready files15 GB
  • FP8 / INT8recommendedweights 6.5 GB7.6 GB
  • 4-bitweights 3.6 GB4.5 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 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.

    # nishankgite/towel-fold-flux3-action-ee50
    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 nishankgite/towel-fold-flux3-action-ee50 --local-dir ~/models/towel-fold-flux3-action-ee50
    # 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 flux-3-action-base (quants, finetunes, LoRAs) will show up under the bell.

Fine-tunes 12All on Hugging Face
LoRAs and adapters 2All on Hugging Face

About the model

nishankgite/towel-fold-flux3-action-ee50
Author
nishankgite
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

300

Ask AI about this model

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

Comments