How do I connect this MCP to my LiteLLM instance?

You need to provide your LiteLLM API URL and your Master Key during the Vinkius setup process. Once entered, the connection is ready to use in your AI client.

Can I use this with Cursor or VS Code?

Yes, this MCP works with any MCP-compatible client, including Cursor, Windsurf, and VS Code.

How does it help with cost management?

The MCP allows you to create teams with specific budget limits and query the exact USD spent by any user or team.

Can I change model routing without restarting the proxy?

Yes, you can use the capability to inject new routing endpoints or delete existing ones dynamically.

Is it possible to isolate different teams?

Yes, you can use the create_team capability to establish organizational isolation and track costs per division.

Can I check the budget and rate limits for a specific proxy key?

Yes. Use the get_key_info capability with the specific Key ID. Your agent will retrieve the exact rate limits, budget constraints, and current RPM usage associated with that token.

How do I see the model fallback paths configured in my proxy?

The get_model_info capability allows your agent to extract the global model directory. You'll see the exact fallback chains (e.g., if OpenAI fails, use Anthropic) and the physical endpoints assigned to each model name.

Can my agent create a new team to track specific division costs?

Absolutely. Use the create_team capability and provide a JSON payload defining the team name and optional budget limits. Your agent will provision the new team identity in LiteLLM, allowing for precise organizational cost tracking.

An astronaut overseeing a futuristic data hub

Download PDF Markdown Version

Connect Litellm LLM Proxy Spend Tracking to your AI

Set it up once and it works in ChatGPT, Claude, Cursor, and any other client you love. It all runs safely in the background, so you can relax and focus on what really matters. Control your LLM infrastructure and costs through natural conversation.

Ask AI about this Connector

Developed, maintained, and hosted by Vinkius.

MCP VERIFIED · PRODUCTION READY · VINKIUS GUARANTEED

Waiting for input…

Works with modern AI clients that support MCP, including ChatGPT, Claude, Cursor, and more.

Your AI does the work. We keep it safe and sound.

No other AI tool gives you this. Your workflow runs in the client you already love, your account stays protected, and you decide what the agent can and cannot do. It is the whole reason to make the switch.

Each Connector lives inside our infrastructure, the same place where your workflow runs. The three things below are already part of it, working quietly in the background so you never have to think about them. Nothing to add, nothing to set up, nothing to worry about.

V8 Isolate01

Each Connector runs inside its own sealed environment.

Each Connector runs inside its own sealed V8 runtime — no access to your files, your system, or other servers. A Connector never sees past its own request.

Kill Switch02

One switch stops every agent. Instantly.

One action deactivates every Connector, revokes every token, and ends every session at once. AI Governance keeps the record.

FinOps per User03

Spending limits that pause instead of surprise.

Every user keeps their own connections and credentials, isolated from everyone else. Anything that would cross your spending limit is held for your approval, not billed.

Three controls, built into the layer your AI connects through.

Curated set · 10 capabilities

We ship only the actions that matter.

Capability set01 / 03

01-04

4 capabilities in this set.

Part of 10 available through LiteLLM.

01
Create team
This capability sets up new organizational teams to track specific cost limits for different divisions.
02
Create user
This capability adds new end-user identities to bridge your Vinkius session with proxy logs.
03
Delete key
This capability revokes an existing LLM proxy key immediately.
04
Delete model
This capability removes specific LLM deployments to stop 500 errors in your production routing.

Capability set02 / 03

05-07

3 capabilities in this set.

Part of 10 available through LiteLLM.

05
Get key info
This capability retrieves the configuration and budget constraints for a specific LiteLLM API key.
06
Get model info
This capability pulls the list of endpoints and fallback paths, such as OpenAI to Anthropic.
07
Get team info
This capability fetches the routing logic and bounds for users within a specific Team UUID.

Capability set03 / 03

08-10

3 capabilities in this set.

Part of 10 available through LiteLLM.

08
Get user info
This capability returns the total USD spent by a specific end-user.
09
Create model
This capability injects new routing endpoints, like a fresh Bedrock deployment, into your proxy.
10
Generate key
This capability produces new proxy API keys to isolate different microservices or teams.

Real work

What teams actually run

Microservice Isolation

Generate unique API keys for every new microservice to keep their usage and costs separate.

Budget Control

Set hard monthly limits for specific departments to prevent unexpected overages.

Dynamic Routing

Add new model endpoints to your proxy on the fly to improve reliability.

Cost Auditing

Ask your agent for a breakdown of token spend per user to identify high-cost accounts.

Automate a task

Put LiteLLM to work on one thing.

Step-by-step automations your AI runs through this Connector: the workflow, the prompts to start with, and the exact capabilities behind each move.

One connector, every AI

LiteLLM works with the most popular AI clients.

These are the most popular clients, each with a step-by-step guide: one link, set up once, with governance and visibility built in. And because everything runs on the MCP standard, the same connection also works in any other compatible client — nothing to rebuild.

Building your own app? The connector is yours to use.

You don't need a client to put LiteLLM to work: the same hosted connection plugs into your own applications and agent code, with the same governance on every request. Build with it, chat with it — one connection for both.

Observed, not estimated

946ms average. Fast in production.

LiteLLM is checked daily against the live service.

Daily averagePeak 971ms

Oct 15Today

Fastest day: 855ms
Slowest day: 971ms
14-day trend: Improving-12%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 10 capabilities arrive ready to run.

Preview access · not provider authentication

The vk_preview_* token belongs to Vinkius preview infrastructure. It lets Claude discover and display the capabilities of LiteLLM, so you can see the experience inside your AI.

It does not authenticate your account with LiteLLM. Actions requiring credentials or live account data may not run until you activate the Connector and authorize the service.

LiteLLM Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_km5X3hHTZMwK6l421yD2vEs8yb47UeU5zl6r4SJ6/mcp

Claude Desktop

Follow the steps below to connect in seconds.

1In Claude Desktop, open Settings → Connectors.
2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
3Click Add and start a new chat — LiteLLM capabilities are ready to use.

Configuration · claude_desktop_config.jsonCopy

{
  "mcpServers": {
    "litellm-llm-proxy-spend-tracking-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_km5X3hHTZMwK6l421yD2vEs8yb47UeU5zl6r4SJ6/mcp"
    }
  }
}

Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline

Step-by-step instructions for each client are in the guide. How to connect

Guided setup for Claude? Set up LiteLLM in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work LiteLLM owners hand off.

This MCP is built for technical teams managing large scale LLM deployments through a centralized gateway.

01
Platform Engineers
They use the agent to manage global gateway configurations and audit fallback paths.
02
AI Ops Teams
They use the agent to monitor real-time spending and adjust team budgets.
03
Backend Developers
They use the agent to generate sub-keys and verify model availability in their IDE.

FAQ

Questions LiteLLM owners ask.

01
How do I connect this MCP to my LiteLLM instance?
You need to provide your LiteLLM API URL and your Master Key during the Vinkius setup process. Once entered, the connection is ready to use in your AI client.
02
Can I use this with Cursor or VS Code?
Yes, this MCP works with any MCP-compatible client, including Cursor, Windsurf, and VS Code.
03
How does it help with cost management?
The MCP allows you to create teams with specific budget limits and query the exact USD spent by any user or team.
04
Can I change model routing without restarting the proxy?
Yes, you can use the capability to inject new routing endpoints or delete existing ones dynamically.
05
Is it possible to isolate different teams?
Yes, you can use the create_team capability to establish organizational isolation and track costs per division.
06
Can I check the budget and rate limits for a specific proxy key?
Yes. Use the get_key_info capability with the specific Key ID. Your agent will retrieve the exact rate limits, budget constraints, and current RPM usage associated with that token.
07
How do I see the model fallback paths configured in my proxy?
The get_model_info capability allows your agent to extract the global model directory. You'll see the exact fallback chains (e.g., if OpenAI fails, use Anthropic) and the physical endpoints assigned to each model name.
08
Can my agent create a new team to track specific division costs?
Absolutely. Use the create_team capability and provide a JSON payload defining the team name and optional budget limits. Your agent will provision the new team identity in LiteLLM, allowing for precise organizational cost tracking.

Explore

More in AI Frontier

Connect LiteLLM to Claude, Cursor, ChatGPT & more

Your AI does the work. We keep it safe and sound.

Each Connector runs inside its own sealed environment.

One switch stops every agent. Instantly.

Spending limits that pause instead of surprise.

We ship only the actions that matter.

01-04

Create team

Create user

Delete key

Delete model

05-07

Get key info

Get model info

Get team info

08-10

Get user info

Create model

Generate key

What teams actually run

Microservice Isolation

Budget Control

Dynamic Routing

Cost Auditing

Building your own app? The connector is yours to use.

LiteLLM Connector

Claude Desktop

Platform Engineers

AI Ops Teams

Backend Developers

How do I connect this MCP to my LiteLLM instance?

Can I use this with Cursor or VS Code?

How does it help with cost management?

Can I change model routing without restarting the proxy?

Is it possible to isolate different teams?

Can I check the budget and rate limits for a specific proxy key?

How do I see the model fallback paths configured in my proxy?

Can my agent create a new team to track specific division costs?

Keywords AI AI Connector

Langfuse (LLM Tracing & Evals) AI Connector

Traefik Proxy AI Connector

NVIDIA API Catalog AI Connector

Suggestions

Track-POD AI Connector

Agent Load Balancer Calculator AI Connector

Portkey AI Connector

Kong Gateway AI Connector