zhoumiaosen/groot-n1p5-lora-libero10-gray-nojitter-bs16-60k

robotics model by zhoumiaosen

Older snapshot: check the original source for current information.

From the publisher

Model card & documentation

Source preview

Read the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.

GR00T-N1.5-3B fine-tuned with LeRobot (0.4.1) on grayscale versions of the 379 LIBERO-10 (LIBERO-Long) demonstrations, on a single RTX 4090. This is the final 60,000-step checkpoint. The model expects grayscale camera images: luminance replicated to 3 channels, using the PIL convert("L") formula. At evaluation time, convert the LIBERO env's camera frames to gray before they reach the policy. It is a color-jitter ablation of an RGB-vs-grayscale study. Its recipe is identical to zhoumiaosen/groot-n1p5-lora-libero10-gray-bs16-60k except that training-time color jitter is turned off. Its RGB counterpart is zhoumiaosen/groot-n1p5-lora-libero10-rgb-nojitter-bs16-60k. The question it answers: does grayscale input alone give the lighting robustness that RGB only gets from color jitter? zhoumiaosen/libero10gray: the 379 LIBERO-10 episodes of openvla/modifiedliberorlds, with every image converted to grayscale and then to LeRobot v3.0 at 10 fps. That is 101,469 frames. It is identical frame for frame to zhoumiaosen/libero10rgb, apart from the grayscale conversion. (wrist camera), observation.state (8-D: eef position, axis-angle, 2…

zhoumiaosen/groot-n1p5-lora-libero10-gray-nojitter-bs16-60k on Hugging Face A short preview, not the full document.

Read the full model card ↗ · Preview checked 2026-10-30T11:05:29.654Z

Inside the original model card — Document outline

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Links from the model card

References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.

Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

Model overview

zhoumiaosen/groot-n1p5-lora-libero10-gray-nojitter-bs16-60k is a robotics model repository published by zhoumiaosen on Hugging Face. The source reports the lerobot library.

This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.

Model facts

Task: robotics
Library: lerobot
Recent downloads (30 days): 0
Cumulative likes: 0
Hugging Face trending score: 0
Reported safetensors parameters: 2,413,522,880
Architecture: Not reported
License: apache-2.0
Access: Not gated by Hugging Face
Created: 2026-10-21T06:33:54.000Z
Last modified: 2026-10-21T06:38:08.000Z

Compatibility and lineage

Language coverage was not reported.

Reported base models: nvidia/GR00T-N1.5-3B

Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.

Use and evaluate

Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.

No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.

Model card and usage instructions

Model files and configuration

Community discussion