Integration

Replicate integration

Machine learning model hosting and inference platform. Run thousands of open-source models including image generation, language models, and audio processing. Features simple API access, model versioning, and scalable deployment for building ML-powered applications.

Connect Replicate How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Replicate.Built from the 35 Replicate tools below, on the connection you add once.

Common Replicate use cases

model discovery
model deployment
schema inspection
model selection
cost control
status polling
image generation
usage monitoring
model fine-tuning
model maintenance

What you can do with Replicate

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Model, 7 operations

create_modelCreate modelPOSTCreates a new model. Register a model container for hosting your own AI models on Replicate.
delete_modelDelete modelDELETEDeletes a model permanently. Remove a model and all its versions from Replicate.
get_modelGet modelGETGets model information. Retrieve model details including description, versions, and usage statistics.
get_model_readmeGet model readmeGETGets the README content for a model. Retrieve model documentation and usage instructions.
list_modelsList modelsGETLists public models available on Replicate. Browse the model catalog with sorting and pagination.
search_modelsSearch modelsGETSearches for models, collections, and documentation. Find models by name, description, or keywords.
update_modelUpdate modelPATCHUpdates model metadata. Modify model description, README, GitHub URL, paper URL, weights URL, or license URL.

Prediction, 6 operations

cancel_predictionCancel predictionPOSTCancels a running prediction. Stop in-progress model execution to avoid unnecessary compute costs.
create_deployment_predictionCreate deployment predictionPOSTCreates a prediction on a deployment. Run inference on dedicated hardware with consistent performance.
create_predictionCreate predictionPOSTCreates a prediction to run a model. Execute any open-source AI model with custom inputs and optional webhooks.
create_prediction_officialCreate prediction officialPOSTCreates a prediction for an official model. Run official models without specifying version ID.
get_predictionGet predictionGETGets prediction status and results. Poll for completion status and retrieve model output when finished.
list_predictionsList predictionsGETLists all predictions. View prediction history with status, timestamps, and model information.

Deployment, 5 operations

create_deploymentCreate deploymentPOSTCreates a new deployment. Deploy a model with dedicated hardware for consistent performance.
delete_deploymentDelete deploymentDELETEDeletes a deployment. Remove a deployment and release its dedicated hardware.
get_deploymentGet deploymentGETGets deployment details. Retrieve deployment configuration, status, and scaling settings.
list_deploymentsList deploymentsGETLists all deployments. View all your model deployments with their status and configuration.
update_deploymentUpdate deploymentPATCHUpdates a deployment. Modify model version, hardware, or scaling configuration.

File, 4 operations

delete_fileDelete fileDELETEDeletes an uploaded file. Remove a file from Replicate storage.
get_fileGet fileGETGets file details. Retrieve metadata about an uploaded file.
list_filesList filesGETLists uploaded files. View all files you have uploaded to Replicate for use in predictions.
upload_fileUpload filePOSTUploads a file. Store files on Replicate for use as model inputs.

Training, 4 operations

cancel_trainingCancel trainingPOSTCancels a training job. Stop an in-progress training to avoid unnecessary compute costs.
create_trainingCreate trainingPOSTCreates a training job. Fine-tune or train models on custom data using trainable model versions.
get_trainingGet trainingGETGets training job status. Monitor training progress, view logs, and retrieve trained model version.
list_trainingsList trainingsGETLists all training jobs. View training history with status, timestamps, and results.

Version, 3 operations

delete_model_versionDelete model versionDELETEDeletes a model version. Remove a specific version from a model.
get_model_versionGet model versionGETGets specific model version details. Retrieve version schema, inputs, outputs, and configuration.
list_model_versionsList model versionsGETLists model versions. View all available versions of a model with creation dates and IDs.

Collection, 2 operations

get_collectionGet collectionGETGets collection details. Retrieve specific collection with full list of included models.
list_collectionsList collectionsGETLists model collections. Browse curated groups of models organized by category or use case.

Account, 1 operation

get_accountGet accountGETGets account information. Retrieve your Replicate account details and usage.

Example, 1 operation

list_model_examplesList model examplesGETLists example predictions for a model. View sample inputs and outputs to understand model capabilities.

Hardware, 1 operation

list_hardwareList hardwareGETLists available hardware. View compute options including CPU and GPU tiers for model deployment.

Webhook, 1 operation

get_webhook_secretGet webhook secretGETGets the webhook signing secret. Retrieve the secret used to verify webhook signatures.

Use Replicate from Claude Code, Cursor or Codex

Connect Replicate in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Replicate operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Replicate

Create your workspace
Sign up free. Nothing to install.
Connect Replicate once
Add your Replicate connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Replicate integration FAQ

What can you do with the Replicate integration?

LiveContext exposes 35 operations on Replicate, covering model, prediction, deployment and file. Common uses include model discovery, model deployment and schema inspection.