Replicate
Replicate allows you to run machine learning models in the cloud via API.
Prerequisites
- A Replicate account
- An API key from Replicate tokens page
Connect
conjra add replicateMCP Tools
run_replicate_model
Run a model on Replicate.
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | In owner/name format |
version | string | No | Specific version hash |
input | object | Yes | Model-specific parameters |
webhook | string | No | Webhook URL for results |
Example: "use conjra to run stable diffusion with prompt a cat"
get_replicate_prediction
Get the status and result of a Replicate prediction.
| Parameter | Type | Required | Description |
|---|---|---|---|
predictionId | string | Yes | Prediction ID |
list_replicate_models
Search for models on Replicate.
| Parameter | Type | Required | Description |
|---|---|---|---|
query | string | No | Search query |
limit | number | No | Results limit (default 20) |