Caila Predict API
To access a specific service, you will need:
- The service identifier, for example: just-ai/openai-proxy. The identifier always consists of two parts:
<author>/<service>. You can view the service identifier on the Catalogue page or you can copy the last two elements from the URL of the service page: https://caila.io/catalog/just-ai/openai-proxy - The API key Keys are created in the Workspace: https://caila.io/workspace/api-tokens
- The type of request and response data, as well as examples of request. This information is usually provided on the service page in the Catalogue.
There are two main methods for accessing services via HTTP:
- predict
- predict-with-config
Structure of the predict request
curl --request POST 'https://caila.io/api/mlpgate/account/{author}/model/{service}/predict' \
--header 'MLP-API-KEY: {api_token}' \
--header 'Content-Type: application/json' \
--data-raw '{predict-request-json}'