Moderations

POST /v1/moderations is OpenAI's moderation endpoint. The model is <provider>/<model>, for example openai/omni-moderation-latest or mistral/mistral-moderation-latest.

cURLcurl

1curl https://api.aile.sh/v1/moderations \
2-H "Authorization: Bearer $AILE_KEY" \
3-H "Content-Type: application/json" \
4-d '{ "model": "openai/omni-moderation-latest", "input": "some text" }'

Billing

A string is one input. An array of strings is one input per string, up to 2,048.
An array of text / image_url items is one input. Only models that read images accept them, and Mistral models take strings only.
The hold is the inputs sent. You pay for the results returned, never more. A response with no results is a 502 and is not charged.
A provider refusal you can fix (400, 404, 413, 422 …) passes through. One about the lender's account is a 502 saying you were not charged. A 502 or 504 arrives as 503 with the real code in x-aile-status.
A key's pins and the routing headers choose the lender here too. x-aile-provider must match the model's prefix, and x-aile-max-price does not apply to a per-input price.
Where the deployment enables it, moderations also take keyless x402.

Results are yours. Aile does not act on them, and a flagged result never counts against you.