"NVIDIA" matching MCP servers:

An MCP server that provides AI agents free access to NVIDIA's hosted AI models (100+ models) with automatic model routing for tasks like reasoning, coding, vision, embeddings, and translation.

Enables MCP-compatible assistants to inspect and diagnose NVIDIA Shield TV and Kodi over a local network, including redacted logs, playback and add-on checks, settings changes, reversible repairs, backups, and rollback. It runs locally and defaults to read-only until writes or plugin browsing are explicitly enabled.

Provides a compressed knowledge graph of the NVIDIA AI developer stack for deterministic traversal, enabling agents to answer questions about dependencies and prerequisites with minimal tokens.

Enables querying a compressed knowledge graph of the NVIDIA AI developer stack, allowing users to traverse declared dependencies, prerequisites, and relationships across 20 domains via read-only MCP tools like query_ckg, get_prerequisites, and search_concepts.

MCP server that integrates with the Cumulus Linux OS via NVUE API to read network configuration, interfaces, bridges, VRFs, BGP, and other network information.

Turnkey on-premise private Agentic AI on NVIDIA DGX. Query specs, model CapEx ROI & join Partner Program.

Exposes NVIDIA GPU metrics (info, utilization, VRAM, temperature) via MCP tools for real-time querying from AI assistants.

MCP server that delegates bounded subtasks to NVIDIA NIM sub-agents, enabling context offload and parallel fan-out for coding assistants like Claude Code and Cursor.

MCP server exposing the Sovereign AI Blog (sovgrid.org), 44+ articles on self-hosted AI for NVIDIA DGX Spark. Three tools: search_blog, get_article, diagnose_sglang.

Provides AI agents with version-pinned NVIDIA CUDA-Q documentation, API reference, and runnable examples via MCP, using offline SQLite full-text search. It resolves docs to match the installed cudaq package to avoid version skew.

MCP server that vets LLM-emitted shell commands BEFORE execution — detects rm -rf nested deep in chains, package-manager glob removal (apt remove 'nvidia'), dd/mkfs filesystem destruction, chmod 777 / chown -R privilege blast, network-exfil via curl | bash, chained shutdown/reboot, git destructive ops. 30 detection rules across 8 families. Sub-second, local, free, MCP-native.

Enables agents to query GPU availability across SSH hosts, find free GPUs, list hosts, and inspect fleet status using only nvidia-smi on the remote side.

VibOps MCP is the control plane between your AI agents and your GPU infrastructure. 74 tools covering: GPU fleet management (deploy, scale, monitor across NVIDIA, AMD, Intel, AWS, Google, Groq), Agent Infrastructure Control Plane (per-agent GPU cost, budget enforcement, model policies, dependency graph), governance (AI Act, SOC 2, immutable HMAC audit chain), and GPU FinOps (chargeback, waste..)..

Enables natural language control of NVIDIA Isaac Sim through the Model Context Protocol, allowing you to create robots, build scenes, run simulations, and debug physics from any MCP-compatible IDE.

A local text-to-image generation server that enables AI agents to generate images, manage models, and browse history using the Z-Image-Turbo model. It supports hardware acceleration across NVIDIA, Apple Silicon, and AMD GPUs via multiple MCP transport protocols.

Enables AI agents to inspect and free NVIDIA GPU VRAM by managing Ollama models, helping make room for loading new models.

Dual-memory MCP system with PostgreSQL 17 + pgvector + NVIDIA NIM embeddings. 11 tools: search, save, profile, health, auto-repair, stats, tune, dedup, checkpoint, maintenance. Sparse Selective Caching (SSC) with 4-phase progressive retrieval. Bilingual EN/ES.

A lightweight MCP server that provides real-time hardware statistics including CPU, memory, disk, and NVIDIA GPU usage. It enables users to monitor system performance and retrieve comprehensive host machine specifications through a standardized interface.

Enables AI assistants to control NVIDIA Isaac Sim by building scenes, loading assets, operating robots and humans, reading sensors, managing simulation, and creating Action Graphs.