How to Use the NVIDIA NIM MCP in OpenAI Agents SDK

Q: Can an OpenAI Agents SDK agent automatically scale NVIDIA NIM replicas?

Yes, the agent uses nimscalereplicas to adjust container counts. It monitors performance metrics first to decide if a scale-up is required.

Q: Does NVIDIA NIM support local model verification with OpenAI Agents SDK?

The agent runs nimlistmodels to verify which LLMs are active on the local backend. This ensures your pipeline only routes requests to fully loaded model profiles.

Q: How do I debug container errors through this integration?

You run nimgetcontainerlogs to pull stdout streams directly from the container layer. This lets your agent inspect stack traces and identify CUDA execution issues.

Control local GPU loads and scale production containers directly from your OpenAI Agents SDK pipelines.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect NVIDIA NIM MCP to OpenAI Agents SDK

Create your Vinkius account to connect NVIDIA NIM to OpenAI Agents SDK 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 NIM with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Monitor hardware health and container states

Your agent runs `nim_check_health_live` to check if the physical host container orchestrator is responsive. This verification happens before sending heavy batch inference jobs.\n\nIf a host hangs, this MCP Server uses `nim_check_health_ready` to verify that the GPU layers finished loading the model weights. The pipeline avoids sending queries to uninitialized containers.

Automate replica scaling via OpenAI Agents SDK

We use `nim_scale_replicas` to change the active container count based on live load demands. Your agent reads the current traffic patterns and scales the hardware up or down dynamically.\n\nThis prevents GPU memory crashes during traffic spikes. The agent coordinates these scaling events natively, keeping latency low without human intervention.

Analyze live GPU telemetry and memory limits

Running `nim_get_gpu_status` extracts physical VRAM allocation and topological limits directly from your active hardware. The agent reviews these limits to prevent out-of-memory errors before they happen.\n\nTo track performance over time, this MCP Server uses `nim_get_metrics` to pull raw Prometheus metrics from the NIM runtime. Your agent reads this data to flag thermal throttling or queue delays.

Setup guide

Set up NVIDIA NIM MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all NVIDIA NIM tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives NVIDIA NIM tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate NVIDIA NIM tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="NVIDIA NIM Agent",
            instructions="You have access to NVIDIA NIM tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

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 NIM. 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 NIM 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 NIM MCP in OpenAI Agents SDK

Install the SDK and configure the MCP Server connection using the streamable HTTP parameters pointing to Vinkius. The SDK auto-discovers the eight NIM tools, exposing them directly to your agent constructors.

Yes, the agent uses `nim_scale_replicas` to adjust container counts. It monitors performance metrics first to decide if a scale-up is required.

The agent runs `nim_list_models` to verify which LLMs are active on the local backend. This ensures your pipeline only routes requests to fully loaded model profiles.

You run `nim_get_container_logs` to pull stdout streams directly from the container layer. This lets your agent inspect stack traces and identify CUDA execution issues.

Yes, Vinkius runs this MCP Server in an isolated, zero-trust sandbox. Your raw GPU metrics and container logs never leave the secure execution context.

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