Robots Atlas>ROBOTS ATLAS
AI PlatformAI-native

NVIDIA NemoClaw

Open-source NVIDIA stack for running OpenClaw always-on agents more safely inside OpenShell sandboxes with policy-based privacy and security guardrails.

Producer:NVIDIAManaged Cloud · On-Premises · EdgeReleased:Mar 16, 2026
NVIDIA NemoClaw
SDK / Languages
3typescript, javasc…
Robotics-Ready

Description

NVIDIA NemoClaw is an open-source reference stack announced on March 16, 2026 at GTC as part of the broader NVIDIA Agent Toolkit. The platform simplifies deploying always-on OpenClaw agents by wrapping them in isolated OpenShell containers with security and privacy controls. NemoClaw installs the NVIDIA OpenShell runtime — part of NVIDIA Agent Toolkit — which enforces policy-based security guardrails and gives users control over agent behavior and data handling. The platform is available as an early preview (alpha) and is not intended for production use.

MLOps LifecycleMLOps LifecycleFull model lifecycle: registry, feature store, prompt management, monitoring and human-in-the-loop.

0/17 supported

Model Registry

Versioning — model artifact versioning
Approval workflows — approval workflow before production
Immutable artifacts — immutability of stored versions
Lineage tracking — tracking data and model relationships
0 / 4 supported · 4 unsupported hidden

Feature Store

Online serving — real-time feature serving
Offline storage — feature storage for training
Streaming ingestion — streaming ingestion (Kafka, Flink)
0 / 3 supported · 3 unsupported hidden

Prompt Management

Prompt registry — central prompt repository
Versioning — prompt versioning and history
Testing frameworks — A/B testing and prompt evaluation
0 / 3 supported · 3 unsupported hidden

Monitoring

Data drift detection — input data drift detection
Concept drift detection — concept drift detection
Hallucination monitoring — LLM hallucination monitoring
Bias evaluation tools — bias evaluation tooling
0 / 4 supported · 4 unsupported hidden

Human-in-the-Loop

Labeling services — data labeling tools
RLHF workflows — reinforcement learning from human feedback
Manual override — manual override of model decisions
0 / 3 supported · 3 unsupported hidden

SecurityEnterprise SecurityCertifications, access controls and data-protection features essential for corporate deployments and cloud privacy compliance.

Developer EcosystemDeveloper EcosystemDeveloper resources: available SDKs, supported programming languages, and infrastructure features and model-deployment methods.

SDK Languages
TSTypeScriptJSJavaScriptPyPython
API Type
REST
Community & resources
Templates library
Quickstarts
API Reference
Tutorials
$

Pricing & Business ModelPricing & Business ModelBilling models (usage-based, provisioned throughput), resource limits and SLA parameters (uptime, support tiers).

Pricing models

Usage-based

SLA & Support

Community

SourcesDocumentation VaultCentralized hub of links to official sources, technical guides, repositories and release notes.