How to Use the NVIDIA AI MCP in LangChain

Q: How do I configure the NVIDIA AI MCP Server within LangChain?

Install langchain-mcp-adapters and langgraph via pip. Initialize the MultiServerMCPClient with the Vinkius endpoint URL, call client.gettools(), and pass those tools directly to your agent constructor.

Q: Which models can I call using NVIDIA AI and LangChain?

You query the listmodels tool to get a live list of supported models like Llama 3.1 or Mistral Large. Your LangChain agent then passes the selected model name directly into the chatcompletion tool parameter.

Run NVIDIA AI models directly inside your LangChain chains and agents with this dedicated MCP Server.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect NVIDIA AI MCP to LangChain

Create your Vinkius account to connect NVIDIA AI to LangChain and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup NVIDIA AI with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain NVIDIA AI reasoning into multi-step agents

The `ask_question` tool lets your agent query a 405B parameter model to solve complex logic puzzles during a run. You feed the output of this step directly into `generate_code` to write the required scripts based on that reasoning. This setup turns static prompts into dynamic chains. Your LangChain agent decides when to pull in `summarize_text` or run `translate_text` based on what the user throws at it, all tracked via LangSmith.

Build structured database pipelines with this MCP Server

The `text_to_sql` tool translates raw user questions into database queries that your LangChain database chain runs immediately. You don't write custom parsers or regex rules because the model handles the schema mapping directly. Combine this with `get_embeddings` to check if an incoming query matches cached results in your vector store. This MCP Server gives your LangChain agent the exact tools it needs to bridge natural language and structured relational databases.

Track every model call with LangChain observability

The `chat_completion` tool routes your system prompts and message history to Llama or Mistral models on the NVIDIA API Catalog. You configure the client to send these payloads through this MCP Server, which exposes raw execution metrics to your tracing dashboard. When your agent calls `list_models` or `analyze_sentiment`, LangSmith logs the exact latency and token count. You see exactly which step in your chain costs the most, making optimization straightforward.

Setup guide

Set up NVIDIA AI MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes NVIDIA AI tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "nvidia-ai-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent NVIDIA AI transactions"
    })
    print(result["messages"][-1].content)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by NVIDIA. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect NVIDIA AI now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about NVIDIA AI MCP in LangChain

Install `langchain-mcp-adapters` and `langgraph` via pip. Initialize the `MultiServerMCPClient` with the Vinkius endpoint URL, call `client.get_tools()`, and pass those tools directly to your agent constructor.

Yes. LangChain agents use ReAct logic to evaluate the user input and select `text_to_sql` only when the prompt requires database access. The agent inspects the tool schema provided by the MCP Server to make this decision.

You query the `list_models` tool to get a live list of supported models like Llama 3.1 or Mistral Large. Your LangChain agent then passes the selected model name directly into the `chat_completion` tool parameter.

The server is stateless by default to keep execution fast. You use `client.session()` in your LangChain code to maintain persistent context across multiple tool executions.

Vinkius runs the server in an isolated V8 sandbox where your text, code snippets, and SQL queries are processed ephemerally. No data is stored on the host after the tool execution finishes, keeping your inputs private.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript