Content moderation

Screen text inputs against safety policies using moderation endpoints.

Code example

from openai import OpenAI

client = OpenAI(
    base_url="https://api.zorveus.com/v1",
    api_key="zrv_your_api_key",
)

response = client.moderations.create(
    model="openai/text-moderation-latest",
    input="Sample text content for automated compliance screening.",
)

result = response.results[0]
print(f"Flagged: {result.flagged}")
print(f"Categories: {result.categories}")

Use the result

Read results[0].flagged for the provider's overall decision and inspect category fields when your policy needs them. Category names and scores can change between moderation models, so store the model ID with any audit decision.

Moderation is a model call. The request still passes through API-key, model-policy, wallet, cap, and credit checks.

Use GET /v1/models to verify that the moderation model is routable. Do not present a moderation result as a legal or compliance determination.