A robotics workbench for training, testing, serving, and running robot-control policies, including ACT, Diffusion Policy, and SmolVLA. It also provides standard interfaces for robot datasets and policy models.

Use it to convert datasets, train or evaluate robot policies, deploy them for serving, run inference, and inspect available checkpoints.

It gives robot-learning projects one defined way to prepare data and run policies across training, evaluation, and deployment. This reduces the need to connect separate tools for each stage.

lerobot — Skill for Claude Code and Codex by nebius

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add nebius/nebius-physical-ai --skill lerobot

Clone the repo

git clone --depth 1 https://github.com/nebius/nebius-physical-ai

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/lerobot/github.svg)](https://agentmods.dev/skills/nebius/nebius-physical-ai/lerobot)

Your own site

<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/lerobot"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/lerobot/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/lerobot"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/lerobot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 27 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 894 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00027	$0.00894
Opus 5	$0.00014	$0.00447
Sonnet 5	$0.00005	$0.00179
Haiku 4.5	$0.00003	$0.00089

Measured 4d ago against content hash 892a4cc21406, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

lerobot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/tools/lerobot/SKILL.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

LeRobot

LeRobot is the default robot policy training framework. It supports ACT, Diffusion Policy, and SmolVLA (and additional VLAs / world models in 0.6.0).

Use it as the data standard and policy interface layer, not as a managed-service competitor to Hugging Face.

Supported versions

Version	Role	Image tag	Notes
0.5.1	Default	`npa-lerobot:cuda13-b300-0.5.1-sm80-sm90-sm100-sm103-sm120-20260803T034152Z`	Accepted public default; the plain `0.5.1` alias is historical
0.6.0	Additional package support	Operator-built image required	No accepted public image pin/digest; lean extras (`training,evaluation,pusht,libero,diffusion,smolvla`); `--eval_freq` → `--env_eval_freq`

Select the package with --lerobot-version. For serverless training on 0.6.0, supply a validated operator image through train --image; version selection alone does not publish an image. VM deployment installs the selected package and has no --image option. The anonymous 2026-09-05 audit returned 404 MANIFEST_UNKNOWN for the official npa-lerobot:0.6.0 tag. It is outside the current public release plan and must not be treated as an accepted release.

Canonical manifest: npa/src/npa/deploy/lerobot_version_manifest.json.

Upstream release notes: https://huggingface.co/blog/lerobot-release-v060

Interfaces

API:

POST /train
POST /eval
POST /serve
POST /infer
GET /list-checkpoints

CLI:

npa workbench lerobot deploy
npa workbench lerobot deploy --runtime vm --lerobot-version 0.6.0
npa workbench lerobot train
npa workbench lerobot train --runtime serverless --lerobot-version 0.6.0 --image '<validated-operator-image>@sha256:<digest>' ...
npa workbench lerobot eval
npa workbench lerobot serve
npa workbench lerobot infer
npa workbench lerobot list-checkpoints

Build operator/BYOF image variants (official publication has separate gates):

npa/docker/workbench/lerobot/build.sh --registry '<operator-registry>' --all-versions
# or
npa/docker/workbench/lerobot/build.sh --registry '<operator-registry>' --version 0.6.0

Read the full file on GitHub · 80 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

4d ago Changed · +5 lines 892a4cc21406
11d ago First seen · 75 lines · 27 tokens per session scan A a436ce9f81b4

Subscribe to this mod's changes

lerobot is a skill published in the GitHub repository nebius/nebius-physical-ai (28 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 894 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Data and AI Hardware and IoT

spark-environment-setup

Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.

wshobson/agents · 76 tokens

spark-memory-thermal-ops

Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.

wshobson/agents · 59 tokens

spark-training-gotchas

Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.

wshobson/agents · 63 tokens

llama-cpp

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.

davila7/claude-code-templates · 76 tokens

minicpm5-deploy-vllm-ascend

Deploy MiniCPM5-2B with vLLM on Huawei Ascend NPU using vLLM-Ascend. Use when the user mentions vLLM-Ascend, Ascend NPU, Huawei Ascend, CANN, torchnpu, davinci devices, or wants an OpenAI-compatible MiniCPM5 server on Ascend hardware.

OpenBMB/MiniCPM · 87 tokens

amc-run-video-calibration

Calibrates pre-recorded cam.mp4 datasets through the AutoMagicCalib REST API. Use for user-supplied local MP4s; route live RTSP streams to amc-run-rtsp-calibration.

NVIDIA/skills · 57 tokens