An astronaut overseeing a futuristic data hub
Download PDFMarkdown Version

Connect Baseten to ChatGPT, Claude and Gemini

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. Run predictions and audit deployments directly from your agent.

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.

ChatGPTClaudeCursorPerplexityGeminiMicrosoft CopilotRaycastMeta AI

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.

A futuristic MCP client docking bay
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 · 6 capabilities

We ship only the actions that matter.

Capability set01 / 02

01-03

3 capabilities in this set.

Part of 6 available through Baseten.

  1. 01

    Get model

    Call get_model when you need specific details about one particular Baseten model.

  2. 02

    Predict

    Execute a prediction using predict. You just need to provide the tensor shapes or dictionary that matches the deployed instance, and the MCP runs the serverless inference prediction.

  3. 03

    Get deployment

    Use get_deployment to pull explicit, detailed information about a specific running deployment instance.

Capability set02 / 02

04-06

3 capabilities in this set.

Part of 6 available through Baseten.

  1. 04

    List secrets

    Run list_secrets to securely enumerate all active environment secrets mapped inside your isolated orchestration ecosystem, without revealing the actual values.

  2. 05

    List models

    Use list_models to get a comprehensive list of all Baseten managed models currently hosted in your account.

  3. 06

    List deployments

    Run list_deployments to see all active inference bounds that match a specific model you're working with.

Real work

What teams actually run

01

Pre-deployment Audit

Before launching a new model, you can list all managed models and check their configurations to ensure all dependencies are met.

02

Live Performance Testing

You run a prediction using a sample payload to test the model's real-time performance against the deployed GPU weights.

03

Resource Verification

You check the deployment status to verify the replica count and autoscaling rules are set up correctly for production.

04

Security Compliance Check

You list workspace secrets to confirm that sensitive environment variables are correctly provisioned and accessible to the running services.

Automate a task

Put Baseten 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

Baseten 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.

ClaudeClaudeChatGPTChatGPTGeminiGeminiPerplexityPerplexityGrokGrokMicrosoft CopilotMicrosoft CopilotCursorCursorVS CodeVS CodeWindsurfWindsurfJetBrainsJetBrainsClineClineLangChainLangChainVercel AI SDKVercel AI SDKLovableLovableZ.aiZ.aiRaycastRaycastQwen CodeQwen CodeKimi CodeKimi CodeLe ChatLe Chat

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

You don't need a client to put Baseten 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

847ms average. Fast in production.

Baseten is checked daily against the live service.

Daily averagePeak 1169ms
Oct 3Today
Fastest day
703ms
Slowest day
1169ms
14-day trend
Slowing+22%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 6 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 Baseten, so you can see the experience inside your AI.

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

Baseten Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_Iok4x74580t8DfXgX0kQJnXTan6vttePrqZ5XRGU/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Baseten capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "baseten-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_Iok4x74580t8DfXgX0kQJnXTan6vttePrqZ5XRGU/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 Baseten in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work Baseten owners hand off.

This MCP is built for technical roles that manage complex ML infrastructure. If your job involves deploying, testing, or auditing models in production, this capability gives your agent the necessary operational control. It keeps you in your core IDE while handling the entire ML-Ops workflow.

  • 01

    ML Engineers

    Execute test payloads to deployments instantly without needing to spin up local Python notebooks.

  • 02

    DevOps/SREs

    Audit running deployment resources and verify replica states reliably directly from your core IDE.

  • 03

    AI Researchers

    Inspect version schemas and manage inference pipeline architectures quickly to validate new concepts.

FAQ

Questions Baseten owners ask.

  • 01

    Does this MCP require me to use a specific AI client?

    No. You connect your preferred agent, as long as it is MCP-compatible. This allows you to use your existing setup, whether it's Claude, Cursor, or VS Code.

  • 02

    Can I use this to manage models that aren't deployed yet?

    Yes. You can use the list_models capability to see all managed models and get their configurations, even if they aren't currently running a live deployment.

  • 03

    What kind of data can I send for a prediction?

    You must provide tensor shapes or a dictionary that strictly matches the deployed instance. The MCP handles running the serverless inference prediction with that payload.

  • 04

    Is the secret listing secure?

    Yes. The list_secrets capability only enumerates the active environment secrets without showing the actual plaintext values, keeping your credentials secure.

  • 05

    Is this just a directory of Baseten features?

    No. Vinkius hosts and manages this MCP. It's a ready-to-use product that executes the commands directly through your agent, eliminating the need to switch contexts.

  • 06

    Can the AI agent run a prediction directly against my hosted model?

    Yes. By pushing a correctly formatted JSON payload to the 'predict' capability, the agent securely triggers inference on the GPU instances, returning the exact calculated response data transparently to your editor context.

  • 07

    Is my workspace and environmental secret data kept safe?

    Baseten secret fetching natively obscures variable values. When you use 'list_secrets', the agent simply evaluates the key names and identifiers existing across your environment to verify configurations without exposing plaintext passwords.

  • 08

    How do I check auto-scaling configurations for an explicitly deployed model?

    You can examine exactly how instances are managed by using 'get_deployment'. Tell the agent to target an active deployment ID and it maps the scaling limits, replica status, and container bounds out-of-the-box.

Explore

More in AI Frontier

Suggestions

Connect Baseten to Claude, Cursor, ChatGPT & more