Use Replicate with your AI.
Connect your account once and let the AI you already use work with it, without building another integration. Automate machine learning workflows via Replicate. run models, manage predictions, and search for AI assets directly from any AI agent.
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.
Complete set · 20 capabilities
The complete Replicate capability set.
These are the exact actions your AI can choose when you ask it to work with Replicate.
01-04
4 capabilities in this set.
Part of 20 available through Replicate.
- 01
Create model
Create a new model on Replicate
- 02
Delete model version
Delete a specific model version
- 03
Get account
Get details about the authenticated user or organization
- 04
Get model version
Get details of a specific model version, including its OpenAPI schema
05-08
4 capabilities in this set.
Part of 20 available through Replicate.
- 05
Get prediction
Get the status and output of a prediction
- 06
List predictions
List recent predictions
- 07
Get model
Get details of a specific model
- 08
Cancel prediction
Cancel a running prediction
09-12
4 capabilities in this set.
Part of 20 available through Replicate.
- 09
Create deployment
Create a new private deployment with custom autoscaling
- 10
Create deployment prediction
Create a prediction using a dedicated deployment
- 11
Create prediction
Create a new prediction to run a model
- 12
Create training
Create a training to fine-tune an existing model
13-16
4 capabilities in this set.
Part of 20 available through Replicate.
- 13
Get collection
Get details of a specific model collection
- 14
Get training
Get the status of a training
- 15
Get webhook secret
Get the default webhook secret key for verifying signatures
- 16
List collections
List curated collections of models
17-20
4 capabilities in this set.
Part of 20 available through Replicate.
- 17
List hardware
List available hardware SKUs and their descriptions
- 18
List model versions
List all versions of a model
- 19
Search models
Search for public models on Replicate
- 20
Update model
Update metadata for an existing model
Observed, not estimated
998ms average. Fast in production.
Replicate is checked daily against the live service.
- Fastest day
- 853ms
- Slowest day
- 1274ms
- 14-day trend
- Stable+2%
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.
https://edge.vinkius.com/vk_preview_ULeVbcRjoLCkIr9x93q2HML8VJd4AXmk4CWHHPso/mcpClaude 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 — Replicate capabilities are ready to use.
{
"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
FAQ
Questions Replicate owners ask.
- 01
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.
- 02
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.
- 03
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
Replicate AI Connector
Run ML models via Replicate — generate images, text, audio and video from community models, track predictions
ViewReplicate AI Connector
Equip your AI to dynamically search, run, and monitor thousands of open-source machine learning models hosted
ViewMetatext AI Connector
No-code NLP and AI model management via Metatext — run inference and manage datasets.
ViewBaseten AI Connector
Manage your Baseten AI models — orchestrate deployments, list secrets, and run serverless inference prediction
View
Suggestions
Hugging Face AI Connector
Access thousands of pre-trained AI models for NLP, vision, and audio tasks with the largest open-source machin
ViewMLflow (ML Lifecycle Management) AI Connector
Manage ML lifecycle via MLflow — track training runs, monitor metrics, and audit the model registry.
ViewModelbit (ML Model Deployments) AI Connector
Deploy and call machine learning models directly from your AI agent using Modelbit's inference endpoints.
ViewDeep Talk AI Connector
Equip your AI agent to analyze conversation datasets, extract topics, and monitor sentiment via the Deep Talk
View
