How to Use the Metatext MCP in LangChain

Q: Can a LangChain agent choose between different Metatext models?

Yes. Your agent can run searchnlpmodels to find active classifiers or extractors, inspect their metadata, and then call runmodelinference on the specific model that matches the current task.

Run NLP predictions and manage datasets directly inside your LangChain reasoning loops.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Metatext MCP to LangChain

Create your Vinkius account to connect Metatext 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 Metatext with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Run Metatext inference inside LangChain chains

The `run_model_inference` tool lets your LangChain agent run text predictions using your deployed NLP models. Instead of writing custom API integration code for your custom classifiers or extraction models, you just pass this tool to your agent. It handles the input formatting, runs the model, and passes the prediction directly to the next step in your chain. You can track the exact inputs, outputs, and latency of these model runs through LangSmith. If a model output doesn't match what the next tool expects, your agent can catch the error and run a different model to fix it.

Find and inspect models dynamically

The `list_nlp_models` and `search_nlp_models` tools give your agent the ability to discover which models are available on your account via this MCP server. When a new task comes in, the agent searches your model list to find the best fit rather than relying on hardcoded model IDs. It can then pull the exact model configuration using `get_model_details` to verify the expected input schema. This dynamic discovery means you don't have to redeploy your LangChain code when you train a new model version. Your agent simply scans the active deployments and routes the traffic to the newest setup on the fly.

Feed training data back into your pipelines

The `create_dataset_record` tool writes new text samples directly into your Metatext training datasets. When your LangChain agent identifies high-confidence predictions or user corrections, it logs them back to your training set instantly. This sets up a continuous feedback loop where your production logs feed your next training run. You can also use `list_dataset_records` to let your agent inspect existing training data before adding a duplicate. It keeps your datasets clean and organized without manual intervention.

Setup guide

Set up Metatext 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 Metatext 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({
    "metatext-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 Metatext 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 Metatext. 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 Metatext 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 Metatext MCP in LangChain

You set your Metatext API token once in the Vinkius platform. The MCP server handles the handshake automatically, so your LangChain adapter configuration stays clean and keyless.

Yes. Your agent can run `search_nlp_models` to find active classifiers or extractors, inspect their metadata, and then call `run_model_inference` on the specific model that matches the current task.

The MCP server runs inside a V8 isolate sandbox on Vinkius, which scales to handle concurrent requests. Your LangChain chains can trigger multiple parallel predictions without hitting local processing bottlenecks.

Yes. Because this is a standard MCP tool integration, every call to `run_model_inference` or `list_nlp_datasets` shows up in your LangSmith dashboard with full payload details and execution times.

Yes. Your dataset records and inference text strings are processed inside ephemeral V8 sandboxes. Vinkius does not store your payload data, and all communication with the Metatext API uses secure TLS encryption.

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