POST /v1/moderations
Request Body
string | array
required
The text to classify. Can be a single string or an array of strings.
string
default:"text-moderation-latest"
The moderation model to use. Options:
text-moderation-latest, text-moderation-stable.Response
string
Unique identifier for the moderation request.
string
The model used for classification.
array
Categories
| Category | Description |
|---|---|
hate | Content expressing hatred toward a group |
hate/threatening | Hateful content that includes threats |
harassment | Content that harasses an individual |
harassment/threatening | Harassment that includes threats |
self-harm | Content promoting self-harm |
self-harm/intent | Content expressing intent to self-harm |
self-harm/instructions | Instructions for self-harm |
sexual | Sexual content |
sexual/minors | Sexual content involving minors |
violence | Violent content |
violence/graphic | Graphic violent content |
Examples
Basic Moderation
from openai import OpenAI
client = OpenAI(
api_key="sk-voidai-your_key_here",
base_url="https://api.voidai.app/v1"
)
response = client.moderations.create(
input="This is a friendly message about programming."
)
result = response.results[0]
print(f"Flagged: {result.flagged}")
if result.flagged:
for category, flagged in result.categories.items():
if flagged:
score = result.category_scores[category]
print(f" {category}: {score:.4f}")
import OpenAI from 'openai';
const client = new OpenAI({
apiKey: 'sk-voidai-your_key_here',
baseURL: 'https://api.voidai.app/v1'
});
const response = await client.moderations.create({
input: 'This is a friendly message about programming.'
});
const result = response.results[0];
console.log(`Flagged: ${result.flagged}`);
if (result.flagged) {
for (const [category, flagged] of Object.entries(result.categories)) {
if (flagged) {
const score = result.category_scores[category];
console.log(` ${category}: ${score.toFixed(4)}`);
}
}
}
curl https://api.voidai.app/v1/moderations \
-H "Authorization: Bearer sk-voidai-your_key_here" \
-H "Content-Type: application/json" \
-d '{
"input": "This is a friendly message about programming."
}'
Batch Moderation
texts = [
"Hello, how are you today?",
"I love learning new programming languages!",
"Let's build something amazing together."
]
response = client.moderations.create(input=texts)
for i, result in enumerate(response.results):
print(f"Text {i}: {'Flagged' if result.flagged else 'OK'}")
const texts = [
'Hello, how are you today?',
'I love learning new programming languages!',
"Let's build something amazing together."
];
const response = await client.moderations.create({ input: texts });
response.results.forEach((result, i) => {
console.log(`Text ${i}: ${result.flagged ? 'Flagged' : 'OK'}`);
});
Content Filtering Integration
from openai import OpenAI
client = OpenAI(
api_key="sk-voidai-your_key_here",
base_url="https://api.voidai.app/v1"
)
def is_content_safe(text, threshold=0.5):
"""Check if content is safe based on moderation scores."""
response = client.moderations.create(input=text)
result = response.results[0]
if result.flagged:
return False
# Check if any category score exceeds threshold
for category, score in result.category_scores.items():
if score > threshold:
return False
return True
def safe_chat(user_message):
"""Chat with content filtering."""
# Check user input
if not is_content_safe(user_message):
return "I'm sorry, I can't respond to that message."
# Get AI response
response = client.chat.completions.create(
model="gpt-4",
messages=[{"role": "user", "content": user_message}]
)
ai_message = response.choices[0].message.content
# Check AI output
if not is_content_safe(ai_message):
return "I generated an inappropriate response. Please try again."
return ai_message
# Usage
result = safe_chat("Tell me about machine learning")
print(result)
Response Example
{
"id": "modr-abc123",
"model": "text-moderation-latest",
"results": [
{
"flagged": false,
"categories": {
"hate": false,
"hate/threatening": false,
"harassment": false,
"harassment/threatening": false,
"self-harm": false,
"self-harm/intent": false,
"self-harm/instructions": false,
"sexual": false,
"sexual/minors": false,
"violence": false,
"violence/graphic": false
},
"category_scores": {
"hate": 0.00001,
"hate/threatening": 0.000001,
"harassment": 0.00002,
"harassment/threatening": 0.000001,
"self-harm": 0.000001,
"self-harm/intent": 0.000001,
"self-harm/instructions": 0.000001,
"sexual": 0.00001,
"sexual/minors": 0.000001,
"violence": 0.00003,
"violence/graphic": 0.000001
}
}
]
}
The moderation API is designed to help filter content but should not be the only safety measure in your application. Consider implementing additional safeguards for production systems.