Skip to content
Robotics

LangGap InternW0-Delta with LDS: Policy for Robotics Tasks Based on InternRobotics

1 min readupdated

This model is a trained robotics policy checkpoint created using the Linear Decodability Supervision method at step 1469. It was trained on all available localization scenes and is designed to work with the native InternW0-Delta LIBERO runtime. The release includes model weights, dataset statistics, and configuration files for deployment. The model will be useful for researchers and developers in robotics and manipulation control.

Key features

  • Uses the Linear Decodability Supervision approach for control policy training
  • Trained on the complete set of localization scenes without splitting into held-out samples
  • Full compatibility with the InternRobotics InternW0-Delta LIBERO ecosystem and runtime
  • Includes all necessary configuration files and dataset statistics for launching

What you can do

  • Run control policy evaluation in the LIBERO environment using the internw0_delta_libero.pt file
  • Integrate weights into the data processing and robotics task simulation pipeline
  • Adjust configuration using the included native_config.yaml file and local assets
  • Verify the integrity of downloaded files using SHA256 checksums

Example

To run the model, load the internw0_delta_libero.pt checkpoint in the upstream InternW0-Delta LIBERO runtime using the accompanying dataset_stats.json and native_config.yaml.

Read moreDetails
Base model: InternRobotics/InternW0-Delta-Libero
Task: robotics
Frozen components: Wan-AI/Wan2.2-TI2V-5B and Alibaba-DAMO-Academy/RynnBrain1.1-2B
Auxiliary LDS heads are dropped during inference
Release contains weights, inference parameters, and stateless configs without optimizer states
Discuss Telegram 0

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 12 GB · ready files13 GB
  • FP8 / INT8recommendedweights 5.8 GB6.9 GB
  • 4-bitweights 3.3 GB4.1 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.

    # RuFlect/LangGap-InternW0-Delta-LDS
    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 RuFlect/LangGap-InternW0-Delta-LDS --local-dir ~/models/LangGap-InternW0-Delta-LDS
    # 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 InternW0-Delta-Libero (quants, finetunes, LoRAs) will show up under the bell.

Fine-tunes 2All on Hugging Face

About the model

RuFlect/LangGap-InternW0-Delta-LDS
Author
RuFlect
Category
Robotics
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

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

Comments