Replicate

liveAI/MLTechnologyDeveloper

Replicate API — browse, run, and manage AI model predictions and deployments across thousands of hosted ML models.

16tools
0msauth
free tier50 calls/day

Tools

list_models

List all public models available on Replicate, paginated via cursor; returns owner, name, description, run count, and latest version id for each model.

No parameters required.

▶Try it
model

Fetch metadata for a specific Replicate model by owner and name, including description, visibility, run count, latest version, and GitHub URL.

No parameters required.

▶Try it
model_versions

List all published versions of a Replicate model by owner and name; returns version ids, creation timestamps, and input/output schema for each version.

No parameters required.

▶Try it
model_version

Fetch full metadata for a specific Replicate model version by owner, name, and version_id, including the complete input/output OpenAPI schema and Cog version.

No parameters required.

▶Try it
search_models

Search public Replicate models by keyword query; returns matching models with owner, name, description, and run count.

No parameters required.

▶Try it
collections

List all curated Replicate model collections (e.g. "text-to-image", "speech-to-text"); returns collection slug, name, and description.

No parameters required.

▶Try it
collection

Fetch models belonging to a specific Replicate collection by slug (e.g. "text-to-image"); returns the collection description and a list of model objects.

No parameters required.

▶Try it
create_prediction

Start a Replicate model run by submitting a version id and input object; returns a prediction id and status that can be polled with the prediction tool until output is ready.

No parameters required.

▶Try it
prediction

Poll the status and retrieve output for a Replicate prediction by id; returns status (starting/processing/succeeded/failed/canceled), output URLs, and logs.

No parameters required.

▶Try it
list_predictions

List your recent Replicate predictions in reverse chronological order, paginated via cursor; returns prediction id, model version, status, and creation timestamp.

No parameters required.

▶Try it
cancel_prediction

Cancel a prediction.

No parameters required.

▶Try it
list_deployments

List your Replicate model deployments (dedicated, auto-scaled endpoints), paginated via cursor; returns owner, name, current model version, and scaling config.

No parameters required.

▶Try it
deployment

Deployment detail.

No parameters required.

▶Try it
create_deployment_prediction

Run via a deployment.

No parameters required.

▶Try it
account

Fetch your Replicate account details including username, name, GitHub URL, and account type.

No parameters required.

▶Try it
hardware

List available Replicate hardware types (CPU, GPU SKUs) with their names and SKU identifiers; use to select hardware when creating a deployment.

No parameters required.

▶Try it

Test with curl

The gateway speaks JSON-RPC 2.0 over HTTP POST. You can test any pack directly from the terminal.

List available tools
bash
curl -X POST https://gateway.pipeworx.io/replicate/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
Call a tool
bash
curl -X POST https://gateway.pipeworx.io/replicate/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_models","arguments":{}}}'

Use with the SDK

Install @pipeworx/sdk to call tools from any TypeScript/Node project.

TypeScript
import { Pipeworx } from '@pipeworx/sdk';
const px = new Pipeworx();
const result = await px.call("list_models", {});
ask_pipeworx
// Or ask in plain English:
const answer = await px.ask("replicate api — browse, run, and manage ai model predictions and deployments across thousands of hosted ml models");