# NVIDIA CUDA Docs MCP

> Lets your coding agent look up NVIDIA's current CUDA documentation and code examples, served by NVIDIA itself.

NVIDIA's hosted CUDA MCP Server, part of Nsight AI. It connects your AI coding agent to current, first-party NVIDIA CUDA documentation and code examples, so the answers your agent gives about CUDA draw on NVIDIA's own docs.

What you can do with it
- Search NVIDIA's CUDA documentation from your agent
- Find NVIDIA's CUDA code examples to build from
- Ask CUDA questions while you write, profile and improve your code, and get answers grounded in NVIDIA-curated docs

How it connects
NVIDIA hosts it at https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs. The first time you connect, you sign in with your NVIDIA Developer account, and your agent reuses that sign-in afterwards. NVIDIA gives setup steps for Claude Code, Codex, Cursor and Antigravity, and any other agent that takes a hosted MCP server address can connect to it.

Good to know
- It's NVIDIA's suggested route for people moving off the Nsight Copilot extension for Visual Studio Code, which NVIDIA is deprecating.
- The hosted server serves NVIDIA-curated documentation. If you work with highly sensitive or proprietary code, NVIDIA suggests its open-source, self-hosted Nsight Copilot Blueprint instead, so your data stays on your own systems.
- It complements your usual debugging and profiling tools rather than replacing them.

- Type: Plugin
- Made by: Nvidia
- Category: Developer tools
- Formats: Remote MCP
- Version: 2026.2.1
- Website: https://developer.nvidia.com/nsight-ai
- Page: https://pluginboard.ai/plugins/nvidia-cuda-docs

## Works with

- ChatGPT
- Claude
- Muse
- Grok Bot
- Gemini
- Grok
- GitHub Copilot
- Cursor
- Claude Code
- Codex
- Antigravity
- Gemini CLI
- Replit
- OpenClaw
- OpenCode
- Hermes Agent

## Install

### ChatGPT

In ChatGPT on the web: Plugins → + → Add custom MCP server, then paste https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs under Connection. Business and Enterprise plans; Pro can add read-only servers.

### Claude

In Claude (web or desktop): Customize → Connectors → Add custom connector, then paste https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs

### Muse

In Muse: In a Muse chat, ask it to create a custom connector for the MCP server at https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs. Then tap Connect on the card it shows and sign in, or add your key.

### Grok Bot

In Grok Bot: In a chat with your Bot, say "Add this MCP server: https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs". It confirms the details with you and sets it up, and the tools are ready from your next message.

### Gemini

In Gemini: At gemini.google.com: Settings → Connected Apps → Add a custom app, then paste https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs

### Grok

In Grok on the web: At grok.com/connectors: New Connector → Custom, then enter https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs and sign in if it asks.

### GitHub Copilot

Add to VS Code: vscode:mcp/install?%7B%22name%22%3A%22nvidia-cuda-docs-mcp%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.copilot.nsight.ngc.nvidia.com%2Fmcp%2Fcuda-docs%22%7D

Copilot CLI:

```shell
copilot mcp add --transport http nvidia-cuda-docs-mcp https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

### Cursor

Add to Cursor: cursor://anysphere.cursor-deeplink/mcp/install?name=nvidia-cuda-docs-mcp&config=eyJ1cmwiOiJodHRwczovL2FwaS5jb3BpbG90Lm5zaWdodC5uZ2MubnZpZGlhLmNvbS9tY3AvY3VkYS1kb2NzIn0%3D

Add to .cursor/mcp.json:

```json
{
  "mcpServers": {
    "nvidia-cuda-docs-mcp": {
      "url": "https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs"
    }
  }
}
```

### Claude Code

Run in your terminal:

```shell
claude mcp add --transport http nvidia-cuda-docs-mcp https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

### Codex

Run in your terminal:

```shell
codex mcp add nvidia-cuda-docs-mcp --url https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

### Antigravity

Add to ~/.gemini/config/mcp_config.json:

```json
{
  "mcpServers": {
    "nvidia-cuda-docs-mcp": {
      "url": "https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs"
    }
  }
}
```

### Gemini CLI

Run in your terminal:

```shell
gemini mcp add --transport http nvidia-cuda-docs-mcp https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

### Replit

In Replit: At replit.com/integrations, open MCP Servers → + Add MCP server, paste https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs (add any headers under Advanced settings), then click Test & save.

### OpenClaw

Run in your terminal:

```shell
openclaw mcp add nvidia-cuda-docs-mcp --url https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs --transport streamable-http
```

If it needs a key or sign-in: For https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs, add --header NAME=value for each header it needs, or --auth oauth and then run openclaw mcp login nvidia-cuda-docs-mcp.

### OpenCode

Run in your terminal:

```shell
opencode mcp add nvidia-cuda-docs-mcp --url https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

If it needs a key: For https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs, add --header 'NAME=value' for each header it needs. Add --global to use it in every project.

### Hermes Agent

Run in your terminal:

```shell
hermes mcp add nvidia-cuda-docs-mcp --url https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs
```

Add to ~/.hermes/config.yaml:

```json
{
  "mcp_servers": {
    "nvidia-cuda-docs-mcp": {
      "url": "https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs"
    }
  }
}
```

If it signs in with OAuth: If https://api.copilot.nsight.ngc.nvidia.com/mcp/cuda-docs asks you to sign in, add --auth oauth to the command, or auth: oauth to its entry.

---
From Pluginboard (https://pluginboard.ai).