Nebius
Official@nebius · Netherlands
Offers specialized infrastructure and orchestration for Physical AI, robotics simulation, and GPU-accelerated world model training on cloud-native clusters.
Agent Skills by Nebius
Showing 57 vetted skills indexed across 2 GitHub repositories.
insights
Aggregate structured reports and manifests into a queryable lineage graph and metrics store.
lerobot
Manage robot policy training, evaluation, and inference with LeRobot.
workbench-tool
Standardize containerized FastAPI services with mandatory health and execution endpoints.
isaac-lab
Orchestrate Isaac Lab reinforcement learning simulations on Nebius GPU clusters.
groot
Manage NVIDIA GR00T model deployment, finetuning, and inference on Nebius clusters.
foxglove
Integrate the Foxglove embedded viewer and MCAP processing tools into NPA agent workflows.
skypilot-workflows
Manage SkyPilot cloud workflow lifecycles on Nebius infrastructure.
dataset
Ingest, validate, and curate sensor data into versioned datasets.
nebius-infra
Provision Nebius cloud resources and manage IAM authentication for physical-AI workflows.
cosmos
Manage Cosmos world model deployment, inference, and serverless training on Nebius.
lichtblick
Serve and visualize robotics log files in a browser.
mjlab
Orchestrate locomotion evaluation and SONIC workbench workflows on Nebius GPU clusters.
soperator
Deploy and manage Slurm-on-Kubernetes clusters on Nebius via declarative YAML.
npa-agent
Orchestrate Nebius Physical AI agent VM deployment and lifecycle operations.
scenario-gen
Generate adversarial scenarios to stress-test physical AI policies.
retargeting
Convert SOMA, G1, and Bones motion artifacts into motion-lib PKL schemas.
fiftyone
Deploy and manage FiftyOne for dataset curation and visualization in physical-AI workflows.
open-dreamer
Orchestrate multi-GPU Open Dreamer world-model training on Nebius clusters.
lancedb
Manage LanceDB vector tables, dataset imports, and materialized views.
sonic
Manages NVIDIA GEAR-SONIC whole-body-control model lifecycle including training, export, evaluation, and serving across GPU clusters.
genesis
Run headless GPU-accelerated RL teacher training on Nebius with Genesis.
oss-solution-registry-onboard
Standardize open-source Physical AI repository onboarding into the NPA registry.
workbench-reference-workflows
Orchestrate physical-AI workflows via SkyPilot YAML templates on Nebius infrastructure.
agent-workflow-operate
Validate NPA agent VM readiness for declarative workflow YAML execution.
Frequently Asked Questions About Nebius
FAQPage SchemaWhat specific tasks can be performed using Nebius infrastructure?▼
Nebius enables the deployment of Physical AI agent VMs, orchestration of multi-GPU world model training, management of robotics simulation artifacts, and execution of Sim2Real reinforcement learning pipelines using declarative configuration files.
Who is the target persona for these technical capabilities?▼
The platform is designed for robotics engineers, machine learning researchers, and infrastructure architects focused on high-performance computing, GPU-accelerated simulation, and the lifecycle management of physical-world models.
What are the primary prerequisites for running workloads on Nebius?▼
Users require access to Nebius cloud resources, configured IAM authentication, and familiarity with declarative YAML specifications for defining cluster jobs, containerized services, and multi-stage pipeline execution.