This guide provides instructions for creating models through the Credo AI API. It assumes the use of environment variables for configuration and relies on Python and the requests library to interact with the API. For more details on the endpoint, see our swagger documentation.
Note:
- Self-hosted customers must replace
https://api.credo.aiwith your Credo AI URL ${TENANT}is the tenant name used to log in to Credo AI- See Authentication for how to get the required
${ACCESS_TOKEN}
Step 1: Create a new model in the model registry
- Body params
name→ Model name must be uniquedescription→ Model description
import requests
url = "https://api.credo.ai/api/v2/${TENANT}/models"
headers = {
"Content-type": "application/vnd.api+json",
"Accept": "application/vnd.api+json",
"Authorization": "Bearer ${ACCESS_TOKEN}"
}
payload = {
"data": {
"attributes": {
"architecture": "string",
"description": "string",
"inputs": "string",
"limitation": "string",
"name": "string",
"outputs": "string",
"performance": "string",
"summary": "string",
"trade_offs": "string"
},
"type": "resource-type"
}
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)