Connect Replicate MCP for AI Agents
Running Generative Media Workflows and ML Model Inference
We take care of the infrastructure, maintenance, security, and governance. Works with:
Claude 
ChatGPT 
Cursor 
Gemini 
Windsurf 
VS Code 
Vercel No credit card required. Experience the power of this integration risk-free.
Waiting for input…
What AI agents can do with Replicate: 12 Tools for Open-Source Model Inference
These tools allow your agent to manage the entire lifecycle of running machine learning models, from discovery and parameter checking to execution and status monitoring.
List models
Retrieves a list of all public machine learning models available on the Replicate platform.
Get account
Fetches details about your authenticated Replicate user account credentials.
List collections
Displays curated groups of related models, like 'Audio Generation' or 'Image-to-Text'.
List deployments
Lists any model deployments you have set up and are actively managing.
Cancel prediction
Stops a prediction job that is currently running before it completes.
Create prediction
Starts a brand new model run by providing the specific model version ID and all required input variables in a structured format.
Get collection
Retrieves a list of models grouped under a specified collection slug, such as 'text-to-image'.
Get model
Gathers detailed information about one specific model to understand its required inputs and parameters.
Get prediction
Checks the current status, output, or error log for a previously started prediction job.
List hardware
Shows you which GPU hardware options are available if you need to run models on specific infrastructure.
List predictions
Generates a log of all the prediction jobs you have recently executed or requested.
Search models
Scans the Replicate platform to find public models that match keywords or use cases you specify.
Frequently Asked Questions
How does Replicate MCP help generate new images from text? +
It lets your AI client run powerful open-source image models on demand. You just describe what you want, and the agent handles finding a suitable model and submitting the job to get the high-quality output.
Can I use Replicate MCP to test different ML algorithms? +
Yes. This MCP is built for testing. You can ask your agent to run predictions on multiple models, allowing you to compare how different open-source algorithms handle the exact same input data.
What if my prediction job fails or gets stuck? +
If a model run stalls or throws an error, your agent can check the status for you using get_prediction and even cancel it immediately with cancel_prediction. You don't have to wait for timeouts.
Is Replicate MCP only for developers? +
No. While it handles technical processes, the user experience is purely conversational. Content creators or non-technical users can delegate complex media tasks—like generating audio—by simply chatting with their agent.
How do I find out which models are available for video? +
You ask your agent to search the catalog, and it will use its internal tools to list model collections or run a targeted search_models query specifically for video generation options.
Can the agent pass a JSON payload directly into a Replicate model? +
Yes. You can utilize the create_prediction action and attach the payload parameter filled out with any required input schema (e.g., specific prompt, num_inference_steps). Since models change inputs constantly, you should always ask your assistant to fetch the schema details first via get_model to verify keys.
Does the prediction command return results instantly? +
No, Replicate's API operates asynchronously. The initial command gives your assistant an ID. You must then ask your AI companion to query the get_prediction tool periodically using that generated ID until it displays the completed status along with the generated web URLs or generated strings.
Can the AI browse trending or curated model collections? +
Yes. Use the list_collections tool to browse curated groups of models organized by category — such as image generation, text-to-speech, or video. Each collection includes a slug and description so you can quickly identify the right set of models for your use case.
Other MCPs in this category
AutoGen
10 toolsOrchestrate Microsoft AutoGen multi-agent workflows. Manage sessions, agent roles, workflows, and monitor execution logs from any AI agent.
LangGraph Cloud (Stateful AI Agents) MCP
10 toolsOrchestrate stateful AI agents via LangGraph Cloud. Manage assistants, monitor conversation threads, and handle human-in-the-loop overrides.
ArcGIS
10 toolsAutomate mapping and spatial analysis via ArcGIS. Perform geocoding, route solving, vehicle routing, and calculate origin-destination matrices from any AI agent.
RunPod MCP Server
7 toolsIntegrate your AI securely to RunPod to cleanly quickly provision scalable GPU pods, manage active instances, and inspect serverless endpoints and custom templates natively.
Related MCPs
Statuspage (Atlassian) MCP Integration
16 toolsManage Atlassian Statuspage. List pages, update configurations, and handle user permissions or access controls via AI.
Cursor Analytics
Cursor Analytics MCP
12 toolsTrack team AI coding metrics from Cursor. Agent edit acceptance rates, tab autocomplete stats, daily active users, model usage, MCP/command/plan/skill adoption, and team leaderboards from any AI agent.
QuickNode MCP Integration
18 toolsManage blockchain infrastructure via QuickNode. Create data streams, configure webhooks, and interact with RPC nodes directly from your AI agent.
CleverTap
8 toolsManage customer engagement and analytics via CleverTap. Track campaigns, monitor user segments, and audit event data directly from any AI agent.
Your AI, connected to everything.
Connect Replicate and 5,600+ more MCP servers to Claude, Cursor, or any AI you use.
No credit card required · Free tier available
