How to Use the Neptune.ai (ML Experiment Tracking) MCP in Pydantic AI

Get type-safe, validated Neptune.ai experiment data in your Python agent with Pydantic AI.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Neptune.ai (ML Experiment Tracking) MCP to Pydantic AI

Create your Vinkius account to connect Neptune.ai (ML Experiment Tracking) to Pydantic AI 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 Neptune.ai (ML Experiment Tracking) with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Reliable Experiment Data, Guaranteed

Here's the main benefit: no more silent failures from unexpected API changes. When your agent calls `search_runs` or `get_attributes`, the response from the Neptune.ai server is automatically parsed and validated against a Pydantic model. You get predictable data structures every time. If a field is missing, a metric is a string instead of a number, or the structure is just wrong, your code gets a `ValidationError`. It fails loudly and immediately. This forces you to build robust agents that don't pass corrupted data downstream.

Model-Agnostic Neptune.ai Access

Pydantic AI doesn't lock you into a single LLM provider. You can write your agent logic once to query your Neptune.ai project, then swap the underlying model as needed. Use OpenAI today, Anthropic tomorrow, or a local model for privacy—it all just works. The tool-calling and validation logic remains the same regardless of which LLM you use. This lets you focus on what your agent needs to do with the Neptune.ai data, not the specifics of each model's API.

Simple, Type-Safe Pydantic AI Setup

Getting started is clean. After a `pip install`, you instantiate an `MCPToolset` with the server URL and pass it to your agent. The tools, like `get_project` and `list_models`, will return Pydantic objects, not raw dictionaries. This means you get full autocompletion in your IDE and type checking from tools like MyPy. You can catch bugs related to data shapes before you even run your agent. It’s about writing correct, maintainable code from the start.

Setup guide

Set up Neptune.ai (ML Experiment Tracking) MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "neptuneai-ml-experiment-tracking-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to Neptune.ai (ML Experiment Tracking) tools.",
)

result = await agent.run("List recent Neptune.ai (ML Experiment Tracking) transactions")
print(result.output)

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 Neptune.ai. 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 Neptune.ai (ML Experiment Tracking) 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 Neptune.ai (ML Experiment Tracking) MCP in Pydantic AI

Every response from this MCP server is automatically validated against a Pydantic model. If the `get_attributes` tool returns data that doesn't match the expected schema, Pydantic AI raises a `ValidationError` instead of letting your agent process bad data.

Yes. Pydantic AI is model-agnostic and works with local model servers like Ollama. This gives you a private, self-hosted way to build agents that can still securely access your Neptune.ai experiment metadata through this MCP connection.

Pydantic AI handles this correctly. Your agent will receive an empty list of the corresponding Pydantic model. Your code won't crash; you'll just get a list with zero items that you can handle gracefully.

Authentication is managed by the Vinkius platform, not your code. You provide a single Vinkius endpoint token to the `MCPToolset`. The server then uses your secured Neptune.ai API token from its environment to talk to the Neptune.ai service.

The server fetches your Neptune.ai metadata—run IDs, parameters, and metrics. Pydantic AI then validates this data on the client side. The MCP server itself runs in a zero-trust, ephemeral sandbox on Vinkius, so your experiment metadata is never stored.

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