An astronaut overseeing a futuristic data hub
Download PDFMarkdown Version

Connect Replicate Extended 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 any model version directly from your AI 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 · 20 capabilities

We ship only the actions that matter.

Capability set01 / 05

01-04

4 capabilities in this set.

Part of 20 available through Replicate.

  1. 01

    Get account

    Retrieves details about your authenticated Replicate user or organization.

  2. 02

    List predictions

    Fetches a list of your recent model predictions.

  3. 03

    Search models

    Searches the Replicate library for public models based on your query.

  4. 04

    Create prediction

    Starts a new prediction for a chosen model.

Capability set02 / 05

05-08

4 capabilities in this set.

Part of 20 available through Replicate.

  1. 05

    Create deployment prediction

    Triggers a prediction using a specific, dedicated deployment.

  2. 06

    Create model

    Registers a new model on your Replicate account.

  3. 07

    Create training

    Initiates a training job to fine-tune an existing model with new data.

  4. 08

    Get model version

    Retrieves details and the OpenAPI schema for a specific model version.

Capability set03 / 05

09-12

4 capabilities in this set.

Part of 20 available through Replicate.

  1. 09

    Get training

    Checks the current progress and status of a training job.

  2. 10

    Get webhook secret

    Gets your default webhook secret key for verifying signature authenticity.

  3. 11

    List collections

    Returns a list of curated model collections available on Replicate.

  4. 12

    List hardware

    Shows the available hardware SKUs and their technical descriptions.

Capability set04 / 05

13-16

4 capabilities in this set.

Part of 20 available through Replicate.

  1. 13

    List model versions

    Lists every version available for a specific model.

  2. 14

    Cancel prediction

    Stops a running prediction using its unique prediction ID.

  3. 15

    Create deployment

    Sets up a new private deployment with custom autoscaling settings.

  4. 16

    Delete model version

    Removes a specific model version from your Replicate account. You'll need to provide the owner, model name, and the version ID.

Capability set05 / 05

17-20

4 capabilities in this set.

Part of 20 available through Replicate.

  1. 17

    Get collection

    Fetches details for a specific curated model collection using its slug.

  2. 18

    Get model

    Gets the full details for a specific model by its owner and name.

  3. 19

    Get prediction

    Checks the current status and retrieves the output of a specific prediction.

  4. 20

    Update model

    Modifies the metadata for an existing model in your account.

Real work

What teams actually run

01

Rapid Prototyping

Test different Stable Diffusion versions by simply describing the image you want to see.

02

Automated Asset Generation

Have your agent generate specific audio or image assets as part of a larger coding or writing task.

03

Model Fine-tuning

Start and monitor training jobs for custom models using your own datasets.

04

Deployment Management

Create and manage private, autoscaling deployments for production-ready model access.

Automate a task

Put Replicate 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

Replicate 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 Replicate 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

1088ms average. Fast in production.

Replicate is checked daily against the live service.

Daily averagePeak 1347ms
Oct 3Today
Fastest day
789ms
Slowest day
1347ms
14-day trend
Slowing+16%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 20 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 Replicate, so you can see the experience inside your AI.

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

Replicate Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_ULeVbcRjoLCkIr9x93q2HML8VJd4AXmk4CWHHPso/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 — Replicate capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "replicate-extended-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_ULeVbcRjoLCkIr9x93q2HML8VJd4AXmk4CWHHPso/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 Replicate in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work Replicate owners hand off.

This MCP is built for technical users who need to bridge the gap between conversational AI and heavy machine learning workloads.

  • 01

    AI Engineers

    They use the MCP to test model versions and inputs without writing boilerplate code.

  • 02

    Creative Technologists

    They trigger media generation models directly within their creative workflows.

  • 03

    Data Scientists

    They monitor training jobs and prediction outputs through a single interface.

FAQ

Questions Replicate owners ask.

  • 01

    How do I connect my Replicate account?

    After subscribing to the MCP on Vinkius, you just need to provide your Replicate API Token to establish the connection.

  • 02

    Which AI clients can I use with this MCP?

    You can use this MCP with any compatible client, including Claude, Cursor, and Windsurf.

  • 03

    Can I run my own fine-tuned models?

    Yes, you can use the training capabilities to create fine-tuned models and then run predictions against them.

  • 04

    Do I need to write code to run a prediction?

    No, you can trigger predictions and manage models using natural language commands through your AI agent.

  • 05

    Can I manage my deployments through this MCP?

    Yes, you can create new private deployments and manage hardware configurations directly.

  • 06

    How can I check if my prediction has finished and see the output?

    Use the get_prediction capability with your Prediction ID. It will return the current status (starting, processing, succeeded, or failed) along with the output URLs or data once completed.

  • 07

    Can I search for specific types of models like 'image-to-text'?

    Yes! Use the search_models capability with your query. It will return a list of public models matching your terms, including their owners and descriptions.

  • 08

    Is it possible to stop a model that is taking too long to run?

    Absolutely. Use the cancel_prediction capability with the target Prediction ID to immediately stop the execution and prevent further usage costs.

Explore

More in Developer Tools

Suggestions

Connect Replicate to Claude, Cursor, ChatGPT & more