Moderations endpoint

[piotrnikel]

Moderation Models enhance CloudFerro Sherlock with robust content filtering capabilities that follow industry-standard implementations.

Expect Llama Guard 8B and a model trained on polish dataset.

These API-compatible moderation tools integrate seamlessly with your existing workflows through the familiar OpenAI-standard /moderations endpoint design.